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 |
|---|---|---|---|---|---|
CLK-2/TEL2 is essential for viability from yeasts to vertebrates , but its essential functions remain ill defined . CLK-2/TEL2 was initially implicated in telomere length regulation in budding yeast , but work in Caenorhabditis elegans has uncovered a function in DNA damage response signalling . Subsequently , DNA damage signalling defects associated with CLK-2/TEL2 have been confirmed in yeast and human cells . The CLK-2/TEL2 interaction with the ATM and ATR DNA damage sensor kinases and its requirement for their stability led to the proposal that CLK-2/TEL2 mutants might phenocopy ATM and/or ATR depletion . We use C . elegans to dissect developmental and cell cycle related roles of CLK-2 . Temperature sensitive ( ts ) clk-2 mutants accumulate genomic instability and show a delay of embryonic cell cycle timing . This delay partially depends on the worm p53 homolog CEP-1 and is rescued by co-depletion of the DNA replication checkpoint proteins ATL-1 ( C . elegans ATR ) and CHK-1 . In addition , clk-2 ts mutants show a spindle orientation defect in the eight cell stages that lead to major cell fate transitions . clk-2 deletion worms progress through embryogenesis and larval development by maternal rescue but become sterile and halt germ cell cycle progression . Unlike ATL-1 depleted germ cells , clk-2–null germ cells do not accumulate DNA double-strand breaks . Rather , clk-2 mutant germ cells arrest with duplicated centrosomes but without mitotic spindles in an early prophase like stage . This germ cell cycle arrest does not depend on cep-1 , the DNA replication , or the spindle checkpoint . Our analysis shows that CLK-2 depletion does not phenocopy PIKK kinase depletion . Rather , we implicate CLK-2 in multiple developmental and cell cycle related processes and show that CLK-2 and ATR have antagonising functions during early C . elegans embryonic development .
CLK-2/TEL2 is a DNA damage checkpoint gene which is essential for viability in budding yeast , C . elegans and vertebrates . DNA damage checkpoints are essential for maintaining genome stability in response to DNA damage and act by coordinating DNA repair and by triggering a transient cell cycle arrest , or apoptosis of affected cells . The loading of a pair of highly conserved PI3 kinase-related kinases ( PIKKs ) , ATM and ATR , to sites of DNA damage acts at the apex of DNA damage response pathways [1] . These kinases have overlapping substrate specificity and phosphorylate multiple targets including the kinases Chk1 and Chk2 [2] , [3] . The first C . elegans clk-2 allele initially referred as rad-5 ( mn159 ) , was isolated in a screen for C . elegans mutants hypersensitive for ionizing irradiation [4] . C . elegans clk-2 temperature sensitive mutants are embryonic lethal at the restrictive temperature of 25°C [5]–[7] . However , the cause of this embryonic lethality is not known . At the “permissive temperature” of 20°C both known clk-2 temperature sensitive alleles lead to a slow growth phenotype that is particularly evident in the clk-2 ( qm37 ) allele , which also shows a reduction in cyclic behaviours such as pharyngeal pumping [5] , [6] . Furthermore , both alleles are defective in various DNA damage responses including DNA damage-induced germ cell apoptosis and cell cycle arrest when propagated at 20°C [5] , [6] . CLK-2/TEL2 has been implicated in S-phase regulation and DNA damage checkpoint responses in fission yeast [8] , [9] , and human CLK-2/TEL2 is required for the DNA replication checkpoint and for DNA crosslink repair [10] . Human and yeast CLK-2/TEL2 directly bind to all PI3K-related protein kinases ( PIKKs ) and are considered to be required for maintaining their stability [8] , [9] . Here we use the C . elegans experimental system to assess the essential functions of CLK-2 during development and cell cycle control . In worms cell cycle progression in early embryos occurs very rapidly , with alternating M and S phases and an apparent lack of gap phases [11] . The timing and pattern of cell division and differentiation is invariant and has been fully characterized [12] . Aberrant embryonic development can therefore be traced by cell lineage analysis and resolved at a cellular level [13] . A relatively high level of DNA damage is tolerated during rapid embryonic cell divisions , possibly as a result of natural selection that favours a rapid pace of replication at the expense of genome integrity [14] . Only high levels of DNA damage or replication failure lead to a DNA damage checkpoint-dependent slowing of cell cycle progression [14] . Interestingly , the DNA damage checkpoint is used during early embryogenesis to contribute to the asymmetry of the first zygotic cell division [15] . In contrast to this , cell proliferation is much slower in the C . elegans germline and DNA damage checkpoint signalling is much more sensitive [11] . The germline is the only proliferative tissue in adult worms . The gonad contains various germ cell types that are arranged in an ordered distal to proximal gradient of differentiation [16] , [17] . The distal end of the gonad is comprised of a mitotic stem cell compartment , which is followed by the transition zone where entry into meiotic prophase occurs . Proximal to the transition zone most germ cells are in meiotic pachytene and subsequently complete meiosis and concomitantly undergoing oogenesis in the proximal gonad . DNA replication failure and DNA double strand breaks lead to a prolonged cell cycle arrest of mitotic germ cells and to apoptosis of meiotic pachytene germ cells [18] . In this DNA damage response pathway CLK-2 and ATL-1 act as upstream DNA damage signalling molecules , while the worm p53 like gene cep-1 is only required for apoptosis [19] , [20] . Thus , CLK-2 and ATL-1 are part of sensitive germ cell DNA damage checkpoint pathways that ensure the faithful transmission of genetic information from one worm generation to the next . C . elegans clk-2 ts mutants show that CLK-2 is required for embryonic development [6] , [7] . As these mutants show an increased level of DNA damage in germ cells at the restrictive temperature , the embryonic lethality might be caused by the accumulation of DNA damage that ultimately may result in the death of the embryo [21] . Given that ATR stability depends on CLK-2 [8] , [9] , the depletion of CLK-2 might phenocopy the atl-1 ( worm ATR ) mutant phenotype , which is germline sterility associated with massive levels of DNA double strand breaks [22] . Furthermore , given that CLK-2 is required for the stability of all PIKKs clk-2 mutations might mimic the phenotype of depleting other PIKKs such as TOR-1 , implicated in nutrient sensing [23] and SMG1 , a kinase involved in nonsense-mediated mRNA decay [24] . Finally , loss of CLK-2 function might result in distinct developmental defects not directly predicted from failing to maintain normal levels of PIKKs or from potential DNA replication and/or DNA damage signalling defects . In this study , we assess the essential defects associated with clk-2 by analysing embryonic cell divisions by cell lineage analysis and by exploiting the C . elegans germline system . We show that clk-2 mutants exhibit defects in early embryonic development and in germline cell cycle progression . These phenotypes do not overlap with reported C . elegans PIKK deletion phenotypes .
We wished to determine why clk-2 mutants fail to complete embryogenesis . We therefore started our analysis by following the embryonic development of the two available recessive clk-2 thermosensitive ( ts ) mutants , mn159 and qm37 , ( Figure S1 ) by cell lineage analysis using 4D microscopy . Analysis of clk-2 mutant lineages at the restrictive temperature of 25°C revealed that asymmetric cell divisions occurred normally during the first three embryonic cell cycles as previously reported [5] , [6] but that cell division timing of all cells was delayed compared to wild type ( Figure 1A , B , Table S1 ) . This delay was more pronounced in clk-2 ( qm37 ) than in clk-2 ( mn159 ) ( Figure 1B , Table S1 ) . In the depicted recordings , the wild type embryo is at the 4-cell stage 11 min after cytokinesis of the P0 cell while the clk-2 ( mn159 ) embryo is about to reach the three cell stage with the AB cell approaching cytokinesis ( Figure 1C ) . The depicted clk-2 ( qm37 ) embryo is at the two cell stage with the AB blastomere just having undergone nuclear envelope breakdown ( Figure 1C ) . Thirty-one minutes after P0 cytokinesis wild type embryos are at the 8-cell stage while both clk-2 mutants are in the 6-cell stage . We next aimed to determine the cause of the cell cycle delay associated with clk-2 mutants . Given that clk-2 ( mn159 ) worms show increased DNA double strand breaks in the mitotic zone of the adult C . elegans germline at the restrictive temperature [22] , we reasoned that the cell cycle delay in clk-2 ( mn159 ) and ( qm37 ) embryos might be due to excessive DNA damage , potentially resulting from compromised DNA replication . We therefore tested whether RAD-51 foci , which are indicative of processed DNA double strand breaks or stalled replication forks [25] , accumulate in clk-2 embryos at the restrictive temperature . We indeed observed increased levels of RAD-51 foci in embryos examined between the 100 and 200 cell stage in both clk-2 ( mn159 ) ( 2 . 14±0 . 62 foci/nucleus n = 7 embryos ) and clk-2 ( qm37 ) ( 0 . 97±0 . 19 foci/nucleus n = 8 ) mutants compared to wild type ( 0 . 2±0 . 02 foci/nucleus n = 6 ) ( Figure 2A ) . These results indicate that clk-2 mutants display a delay in embryonic cell cycle timing and increased genomic instability . Given the delay in cell division timing and the accumulation of RAD-51 foci in clk-2 mutants , we asked if the delay is due to the activation of the DNA replication checkpoint . Previous studies showed that the ATL-1/CHK-1 checkpoint is needed for sensing replication failure in C . elegans embryos [15] . Furthermore , the ATL-1/CHK-1 checkpoint contributes to developmental asymmetry by being in part responsible for the DNA replication delay in the P1 cell . Co-depletion of atl-1 and chk-1 is needed to fully inactivate the DNA replication checkpoint [15] . We observed that upon atl-1/chk-1 depletion cell cycle timing is faster beyond the first embryonic cell division ( Figure 2B , Table S1 ) . We therefore conclude that the ATL-1/CHK-1 pathway acts in normal C . elegans early embryonic development to slow down cell cycle progression . As expected , atl-1/chk-1 ( RNAi ) rescued the prolonged cell cycle delay associated with depleting the DIV-1 DNA polymerase primase alpha-subunit [15] ( Figure 2B , Table S1 ) . Importantly , RNAi-mediated atl-1/chk-1 depletion largely rescued the delay in cell division timing associated with both clk-2 mutants ( Figure 2B , Table S1 ) . Our results thus indicate that clk-2 ( mn159 ) and clk-2 ( qm37 ) mutations result in increased DNA damage , which triggers the ATL-1/CHK-1 checkpoint . It has previously been shown that embryonic lethality associated with dut-1 ( RNAi ) , which leads to the misincorporation of uracil during DNA replication is partially rescued by clk-2 ( RNAi ) and chk-1 ( RNAi ) as well as by the clk-2 ( mn159 ) mutation [26] . These results hint towards a checkpoint function of CLK-2 in embryonic cell divisions . We therefore assessed if CLK-2 functions in DNA damage checkpoint signalling in embryos and asked if the cell cycle delay caused by div-1 ( RNAi ) depends on clk-2 . We found that the delay in S-phase progression of the P1 cell caused by div-1 ( RNAi ) is partially rescued by both clk-2 ts alleles ( Table S2 ) . These results suggest that CLK-2 has a checkpoint function in early embryos . However , the AB cell cycle delay is not rescued likely due to the above described cell cycle delay associated with clk-2 ts mutations . It was reported that CLK-2 and CEP-1 , the single C . elegans p53 homolog , cooperate in pathways leading to germ cell apoptosis upon treatment with ionizing irradiation ( IR ) [19] , [20] . cep-1 mutants are defective in IR induced apoptosis but are wild type for IR induced cell cycle arrest and DNA repair suggesting that CEP-1 acts downstream of CLK-2 in the DNA damage response pathway . Derry et al . also observed that a cep-1 deletion partially rescues the slow growth phenotype associated with clk-2 ( mn159 ) and clk-2 ( qm37 ) [27] . We first confirmed the reported partial rescue of the slow growth phenotype of clk-2 ( mn159 ) and ( qm37 ) by the cep-1 ( lg12501 ) deletion ( Figure S2A/B ) [27] . Given the rescue of the clk-2 slow growth phenotype by cep-1 we wondered if cep-1 ( lg12501 ) would suppress the embryonic cell cycle delay of clk-2 mutants . cep-1 ( lg12501 ) , which results in a slightly slower developmental rate compared to wild type , partially rescued the embryonic cell cycle delay associated with both clk-2 alleles ( Figure 2C , Table S1 ) . In contrast , the cell cycle delay in div-1 embryos was not rescued by cep-1 ( lg12501 ) ( Figure 2C , Table S1 ) . This may indicate that distinct DNA lesions occurring in clk-2 mutant embryos but not a general failure of DNA replication as it occurs in div-1 mutations leads to the activation of a cep-1 dependent checkpoint during early C . elegans embryogenesis . In addition , we found that clk-2 ( mn159 ) or ( qm37 ) ; cep-1 ( lg12501 ) double mutants develop to a later embryonic stage and often arrest in morphogenesis stage , with clear signs of tissue differentiation such as the formation of the pharynx or the appearance of gut granules . This late arrest never occurs in either clk-2 single mutant or atl-1/chk-1 ( RNAi ) depleted clk-2 embryos ( Figure 3A , B ) . Given the rescue of the clk-2 mutant cell cycle delay by a cep-1 deletion , we asked if CEP-1 might be modified in clk-2 mutant worms and assayed for changes in its abundance by western blotting . We found that the levels of CEP-1 protein were markedly increased in extracts prepared from synchronised adult clk-2 ( mn159 ) and clk-2 ( qm37 ) worms compared to wild type , indicating that the checkpoint triggered by clk-2 mutations leads to the accumulation of CEP-1 ( Figure 3C ) . This accumulation of CEP-1 likely results from increased CEP-1 in embryos . CEP-1 germline levels are not increased in clk-2 mutants ( data not shown ) and besides embryonic and germline expression CEP-1 is only expressed in very few cells in the pharynx [19] ( data not shown ) . In summary , we show that deleting cep-1 partially rescues the slow growth phenotype associated with clk-2 mutants and that CEP-1 accumulates in clk-2 mutants . It will be interesting to determine the mechanism of CEP-1 accumulation and if other embryonic defects also lead to CEP-1 accumulation . We speculated that there might also be phenotypes occurring in early clk-2 ( mn159 ) and clk-2 ( qm37 ) embryos that are not linked to the cell cycle delay of clk-2 mutants . Indeed , our lineage analysis revealed that 2 out of 7 clk-2 ( mn159 ) and 6 out of 12 clk-2 ( qm37 ) mutant embryos recorded at 25°C exhibit a distinct lineage defect ( Figure 4 ) . We found an abnormal spindle rotation of the ABar cell ( the anterior right granddaughter of the AB founder cell ) at the 8-cell stage in clk-2 mutants . In the six clk-2 ( qm37 ) embryos showing the abnormal spindle rotation ABar divided on average 48±6° off the a-p axis placing ABarp towards the ventral side of the embryo . In five wild-type embryos ABar divided on average by 54±14° off the a-p axis placing ABarp towards the dorsal side of the embryo . The ABar spindle in the six affected clk-2 ( qm37 ) embryos thus derived 102° from wild type . This abnormal rotation gives rise to mispositioned ABarp and ABara daughters at the 12 cell stage , bringing ABarp instead of ABara in contact to the MS blastomere ( Figure 4 , Videos S1 , S2 , and S3 ) . The MS blastomere emits an inductive signal which in wild type is part of the left versus right cell fate decision ( Figure 4 , arrows , Videos S1 , S2 , and S3 ) [28]–[30] . As a consequence cell fates of the early founder cells are changed in the clk-2 mutants , the ABara and ABarp blastomeres adopt the fates of their left counterparts , ABala and ABalp , respectively ( data not shown ) . This change in cell fate identity leads to embryonic death . A failure of the ABar blastomere to rotate the spindle properly can be taken as an indication that spindles are generally not polarised properly [31] , which is a hallmark of mutants in mom-2 ( wnt ) and mom-5 ( frizzled ) [32] . Future work will reveal , if clk-2 influences the Wnt pathway directly or if the observed clk-2 phenotype is independent of this pathway . To further assess potential developmental and cell proliferation-associated defects of clk-2 mutants , we analysed the germline of clk-2 mutants . clk-2 ts mutants are deficient in responding to DNA damaging agents [5] at the “permissive temperature” of 20°C and shifting clk-2 ( mn159 ) mutants to 25°C at the L4 stage leads to the accumulation of DNA damage in affected germ cells [22] . However , these studies were done with the clk-2 ts alleles . As it is not clear whether they act as null alleles at 25°C we analysed a clk-2 deletion allele . The clk-2 ( tm1528 ) deletion allele provided by the Japanese C . elegans knockout consortium lacks part of the 5′ region , the first three exons , and a part of the fourth exon ( Figure S1A ) . Western blotting with a CLK-2 specific antibody provided by Simon Boulton failed to detect any CLK-2 protein in clk-2 ( tm1528 ) worm extracts ( Figure S1B ) . We found that the major phenotype associated with the clk-2 ( tm1528 ) deletion mutant kept at 20°C is not embryonic lethality but germline sterility ( Figure 5A , see below ) and that the same phenotype occurs when the clk-2 ( tm1528 ) deletion mutant is kept at 25°C ( data not shown ) . The clk-2 ( tm1528 ) phenotype is recessive ( data not shown ) . Given that clk-2 ( tm1528 ) worms go through embryogenesis whereas clk-2 ( mn159 ) and ( qm37 ) worms arrest during embryogenesis at the restrictive temperature , we assume that clk-2 ( tm1528 ) worms are rescued by maternal contribution . To ascertain that the missing embryonic lethality of the clk-2 ( tm1528 ) mutant is indeed caused by the maternal supply we reviewed the phenotype of clk-2 ( mn159 ) and clk-2 ( qm37 ) worms by shifting those mutants to 25°C at the L1 stage . Under these conditions we found that clk-2 ( qm37 ) worms are 100% sterile similar to clk-2 ( tm1528 ) worms , while the weaker allele mn159 does not lead to sterility ( Figure 5A ) . Both ts alleles , as well as the deletion , lead to a protruding vulva phenotype ( pvl ) ( Figure 5A ) . This phenotype is often associated with sterile germlines and general problems in postembryonic cell cycle progression [33] . clk-2 ( qm37 ) and clk-2 ( tm1528 ) gonades are significantly smaller in size than those of wild type and clk-2 ( mn159 ) mutants and clk-2 ( qm37 ) and ( tm1528 ) gonads showed a dramatic reduction of germ cell numbers ( Figure 5B , Figure S3 ) . This reduction in germ cell numbers and germline sterility was also obtained upon clk-2 RNAi in the weaker clk-2 ( mn159 ) mutant , further indicating that the clk-2 ( qm37 ) and clk-2 ( tm1528 ) germline phenotypes represent the clk-2 null phenotype ( Figure 5B ) . These results are in contrast to a previous report which stated that no sterility of clk-2 ( qm37 ) germlines was observed [6] . The reduced germ cell number raised the question whether CLK-2 is required for germ cell proliferation or germ cell differentiation . To address this question we performed a time course analysis of germline development in wild type and clk-2 ( tm1528 ) worms . We found that both strains have similar numbers of germ cells up to the L4 stage at which point germlines of clk-2 ( tm1528 ) worms stop proliferating ( Figure 5C ) . To further assess if this phenotype is caused by a proliferation defect we took advantage of gld-2 ( q497 ) gld-1 ( q485 ) double mutants which have germlines that do not enter meiosis and are thus entirely proliferative . Comparing gld-2 ( q497 ) gld-1 ( q485 ) germlines to gld-2 ( q497 ) gld-1 ( q485 ) ; clk-2 ( tm1528 ) triple mutant germlines we found that germ cell numbers are dramatically reduced in the triple mutant indicating that clk-2 has a role in germ cell proliferation rather than in germ cell differentiation ( Figure 5D ) . In addition , clk-2 ( tm1528 ) and clk-2 ( qm37 ) germ cells are larger than wild type . This phenotype , which is reminiscent of arrested mitotic germ cells after ionizing irradiation , indicates that cells might stop cell division but continue with cellular growth [18] ( Figure 5B , arrowheads ) . In summary , our data suggest that CLK-2 is required for cell cycle progression in germ cells . Given that clk-2 mutations lead to a DNA damage checkpoint dependent delay of embryonic cell cycle progression ( Figure 2B ) and given that clk-2 ( mn159 ) germ cells showed elevated levels of RAD-51 foci indicative of faulty replication when shifted to the restrictive temperature at the L4 stage [22] , we suspected that the germ cell cycle arrest of the clk-2 ( tm1528 ) mutant might be due to the activation of the DNA damage checkpoint . We therefore examined if RAD-51 foci occur in the mitotic compartment of clk-2 ( tm1528 ) germ cells . To our surprise we found that like in wild type germ cells , RAD-51 was mainly localized in the cytoplasm of clk-2 ( tm1528 ) germ cells and did not form nuclear foci ( Figure 6A , Video S4 , Table 1 ) . In contrast , clk-2 ( mn159 ) shifted to the restrictive temperature of 25°C at the L1 or the L4 stage accumulated RAD-51 foci ( Figure 6A , Video S6 , Table 1 ) while clk-2 ( qm37 ) formed fewer foci ( Video S5 , Table 1 ) . Thus RAD-51 foci accumulate mostly in the weak clk-2 ( mn159 ) allele as reported previously [22] , while less foci formation is observed in clk-2 ( qm37 ) and only very few RAD-51 foci can be found in clk-2 ( tm1528 ) ( Table 1 ) . The defect in RAD-51 foci formation in clk-2 ( tm1528 ) might be due to a cell cycle arrest outside of S-phase or due to a failure to process DNA double strand breaks , which is needed for RAD-51 focus formation . To test whether DNA double stand break processing is defective in clk-2 ( tm1528 ) mutants we tested whether focus formation occurred after inducing DNA double strand breaks by exposing worms to ionizing irradiation . Irradiation-induced RAD-51 focus formation indicated that double strand break processing occurs normally in clk-2 ( tm1528 ) worms ( Figure 6B ) . Summing up , these results indicate that the clk-2 ( tm1528 ) deletion does not lead to excessive DNA damage and that CLK-2 is not needed for DNA double strand break processing . To further analyse the cell cycle arrest associated with CLK-2 depletion , we asked if clk-2 ( tm1528 ) germ cells arrest in a distinct cell cycle stage . To facilitate this analysis we first established G2 and M phase cell cycle markers . Prior to mitotic entry Cdk1 is kept inactive by Tyr-15 phosphorylation [34] , [35] . An antibody recognizing Tyr-15 phosphorylation of mammalian Cdk1 cross reacts with the corresponding phospho-epitope of C . elegans NCC-1/CDK-1 . Phospho-NCC-1/CDK-1 can be detected until late prophase in worm embryonic divisions [36] . To confirm that phospho-NCC-1 is indeed indicative of G2/M arrested germ cells , we irradiated wild type germlines and found that all germ cells arrested in G2 with high levels of NCC-1P-Tyr15 ( Figure 7A ) . We observed NCC-1 Tyr-15 phosphorylation in only few wild type and clk-2 ( mn159 ) mitotic germ cells but found that all clk-2 ( tm1528 ) cells and clk-2 ( qm37 ) cells showed high levels of NCC-1 Tyr-15 phosphorylation even in the absence of ionizing irradiation ( Figure 7A ) . The clk-2 prophase arrest phenotype might be caused by a direct prophase defect or alternatively by replication defects which could trigger a checkpoint-dependent late G2/M cell cycle arrest . To assess these possibilities , we depleted atl-1/chk-1 in clk-2 ( tm1528 ) worms . The efficiency of atl-1/chk-1 ( RNAi ) depletion was confirmed by observing germ cell micronuclei [22] and by the embryonic lethality of the progeny of RNAi depleted wild type worms ( data not shown ) . We found that all cells of clk-2 ( tm1528 ) atl-1/chk-1 ( RNAi ) germlines were NCC-1 Tyr-15 phosphorylation-positive ( Figure 7B ) . We therefore conclude that cell cycle arrest is unlikely to be mediated by activation of the ATL-1/CHK-1 DNA damage checkpoint ( Figure 7B ) . To further analyze the cell cycle stage of clk-2 ( tm1528 ) germ cells we also used antibodies against phosphorylated histone H3 ( P-H3 ) . In C . elegans P-H3 staining can be observed in cells from prophase/early prometaphase to late telophase [37] . When wild type gonads were stained with anti-P-H3 antibody only 2–5 nuclei per gonad arm were stained and all stained cells displayed a metaphase-like morphology . While we observed the same phenotype for clk-2 ( mn159 ) worms grown at 25°C , all germ cells were P-H3 positive in clk-2 ( qm37 ) worms propagated at 25°C and in clk-2 ( tm1528 ) worms ( Figure 7C ) . However , P-H3 positive cells did not show a metaphase-like morphology . Rather , in most nuclei chromosomes appear to be partially condensed but not aligned at the metaphase plate suggesting a prophase or very early pro-metaphase arrest . This arrest neither depends on atl-1/chk-1 ( Figure 7D ) , nor on cep-1 ( Figure S4 ) . Thus while cep-1 and atl-1/chk-1 are required for delaying cell cycle progression in clk-2 embryos , the germ cell cycle arrest observed in clk-2 ( tm1528 ) mutants does not depend on either of these genes . Given the early prophase arrest we also assessed centrosome behaviour in clk-2 ( tm1528 ) germ cells . Centrosome duplication occurs during S-phase and centrosomes split during late G2 phase . In prophase , centrosome maturation is an essential prerequisite for the assembly of the mitotic spindle , and centrosomes can be visualized through the accumulation of α and γ-tubulin ( for review see , [38] ) . Increased α-tubulin nucleation is followed by the formation of mitotic spindles [38] . When gonads were immunostained for γ-tubulin [39] to label centrosomes we found that centrosome duplication occurs normally in clk-2 ( tm1528 ) worms ( Figure 8A ) . Furthermore , double immunostaining for γ-tubulin and α-tubulin ( Figure 8B ) showed that several wild type germ cells exhibited accumulated α-tubulin , indicative of centrosome maturation and imminent spindle formation . In contrast , no α-tubulin accumulation and no spindle formation could be observed in clk-2 ( tm1528 ) germ cells , although germ cells with duplicated and separated centrosomes were present ( Figure 8B ) . These results raise the possibility that the prophase-like cell cycle arrest phenotype of clk-2 ( tm1528 ) germ cells might be due to the activation of the spindle assembly checkpoint , which responds to defects in spindle formation and kinetochore-microtubule attachment and blocks anaphase progression until correct bi-orientation has occurred [40] . We therefore tested if the RNAi depletion of the C . elegans MAD1 spindle checkpoint gene ortholog mdf-1 [41] rescues the cell cycle arrest phenotype observed in clk-2 ( tm1528 ) worms . Even though both wild type and clk-2 ( tm1528 ) strains displayed the typical previously described pre-meiotic like morphology of mdf-1 ( RNAi ) germ cells [41] ( Figure 8C ) , mdf-1 ( RNAi ) clk-2 ( tm1528 ) germ cells still uniformly stained P-H3 positive ( Figure 8C ) . In summary , our analysis of clk-2 germlines suggests that clk-2 is essential for cell proliferation and that cells deficient in CLK-2 arrest in prophase without forming a mitotic spindle . The CLK-2 cell cycle arrest phenotype is independent of DNA damage and spindle checkpoint activation .
It has recently been shown that CLK-2/TEL2 interacts with all PIKKs in budding and fission yeast as well as in mammals [8] , [9] , [42] , [43] . CLK-2/TEL2 depletion leads to reduced levels of PIKKs , and using CLK-2/TEL2 mouse knockout lines it was shown that the half life of PIKKs is reduced in those cell lines [9] . This finding together with the notion that PIKK dependent checkpoint signalling is reduced in cells lacking CLK-2/TEL2 led to the hypothesis that CLK-2/TEL2 might function in checkpoint signalling by regulating PIKK kinase levels . Given the conservation of the CLK-2 PIKK interaction it is likely that this interaction also occurs in C . elegans , albeit we could not confirm this since we were unable to generate specific CLK-2 and ATR antibodies suitable for immunoprecipitation from worm extracts ( data not shown ) . Nevertheless , our genetic results suggest that , at least in C . elegans , CLK-2 depletion does not phenocopy PIKK depletion phenotypes ( summarized in Table 2 ) . atl-1/ATR and clk-2 mutations have opposite phenotypes during embryonic development and a clk-2 deletion , unlike atl-1 depletion [22] , does not lead to mitotic germ cell catastrophe . Concerning ATM , this worm PIKK is primarily involved in responding to UV-induced DNA damage where , like CLK-2 it is required for UV-induced apoptosis [44] . Furthermore , an atm-1 deletion only shows weak defects in responding to ionizing irradiation [44] , unlike clk-2 ( qm37 ) and clk-2 ( mn159 ) point mutations . Similarly , clk-2 deleted worms do not resemble worms depleted for tor-1 , which arrest in the L3 larval stage and show concomitant gonadal and intestinal degradation [45] . It is possible that partial tor-1 depletion which results in a slow growth phenotype and enhanced longevity [46] , overlaps with the clk-2 ( qm37 ) phenotypes that include a slow growth and a relatively weak longevity phenotype [6] , [7] , [21] . However , the enhanced life span of clk-2 ( qm37 ) worms is rather weak and clk-2 ( tm1528 ) life span is dramatically reduced compared to wild type ( data not shown ) . Our evidence suggesting that CLK-2/TEL2 might not predominately act by regulating PIKK stability is also supported by recent evidence from the budding yeast system . While steady state levels of the budding yeast ATR homologue TEL1 are somewhat reduced in tel2-1 mutants , it was shown that TEL2 is required for the loading of TEL1 to sites of DNA damage [47] . In addition , the finding that TEL2 binding to the budding yeast MEC1 ATM like kinase is lost in tel2-1 mutants while MEC-1 remains functionally intact [48] , points towards the possibility that CLK-2 be able to regulate ATM and ATR PIKKs by mechanisms not directly related to TEL2 PIKK interaction . We observed that cell cycle progression in early clk-2 mutant embryos is generally delayed and is associated with DNA damage accumulation ( for summary see Table 2 ) . The clk-2 cell cycle delay is partially suppressed by atl-1/chk-1 and cep-1 deletion . These results are surprising in the light of previous reports suggesting that CLK-2 and ATL-1 might act together in C . elegans DNA damage response signalling in germ cells [22] . These two proteins might thus act in different pathways during C . elegans embryogenesis . Our results suggest that ATL-1 is active in clk-2 ts mutants . Thus even if there is a reduced level of ATL-1 protein in clk-2 mutant worms , enough ATL-1 is left to cause a cell cycle delay . In embryos depleted for DNA replication factors cell cycle progression is delayed starting from the very first cell cycle and upon division of the zygote the posterior daughter , referred to as the P1 cell , is particularly strongly affected [49] . This delay depends on the ATL-1/CHK-1 dependent DNA damage checkpoint . The relatively weak replication defect of CLK-2 could be due to partial loss of function in the clk-2 ( mn159 ) or ( qm37 ) point mutants or due to CLK-2 being required for faithful DNA replication rather than replication per se . Our genetic analysis implicates the C . elegans p53-like gene cep-1 in the cell cycle delay associated with clk-2 mutants during embryonic cell divisions . Interestingly , deleting cep-1 alleviates the cell cycle delay of clk-2 mutants but does not have an effect on the delay caused by div-1 mutants . Thus distinct DNA replication defects caused by div-1 and clk-2 depletion might lead to differential checkpoint activation . Our results implicate cep-1 in an embryonic DNA integrity checkpoint . Future studies will be required to address how cep-1 can slow embryonic cell cycle progression and which exact replication defects trigger CEP-1 accumulation . Despite a possible role of clk-2 in embryonic DNA replication , clk-2 ( tm1528 ) germ cells still undergo replication and do not display overt signs of genome instability . Analysis of clk-2 ( tm1528 ) deletion mutants reveals that these worms progress through embryogenesis due to maternal rescue but then halt cell cycle progression in the germline . This arrest is distinct from the cell cycle arrest induced by DNA damage and does not require the ATL-1/CHK-1 DNA damage checkpoint and CEP-1 . Similarly , this arrest does not require the spindle checkpoint . It will be interesting to assess if the cell cycle arrest is due to the requirement of clk-2 in G2 cell cycle progression or due to the activation of a further checkpoint such as the p38 stress activated checkpoint [50] . clk-2 ( tm1528 ) worms arrest in a phospho-histone H3 positive pro-metaphase like stage with partially condensed chromosomes , while DNA damage leads to a G2 arrest characterized by high levels of phosphorylated CDK-1 Tyr 15 and the absence of phosphorylated histone H3 in wild type worms . Interestingly , CDK-1 Tyr 15 is still phosphorylated in these arrested germ cells , indicating that these cells arrest with low CDK-1 activity . Thus our data suggest that there might be an uncoupling of mitotic events in clk-2 ( tm1528 ) germ cells . Clk2/Tel2 has also been shown to be required for cellular proliferation in mouse embryonic fibroblasts . The arrest after CLK-2/TEL2 depletion is not uniform in TEL2 deficient MEFs . These cells arrest with an increased proportion of cells with a 2N or 4N DNA content , and a reduced S and M phase index , and were reported to show a ‘senescence-like flattened morphology’ [8] . Thus CLK-2/TEL2 might have additional functions in mammalian cells that are not directly related to cell cycle regulation . Alternatively , a cell cycle regulatory function of CLK-2/TEL2 might not be uniformly needed in all cell types . Our analysis of clk-2 mutant phenotypes reveals distinct CLK-2 functions in embryonic cell cycle progression and in germ cell cycle progression . The clk-2 ( tm1528 ) null allele results in the most severe germline phenotype . At present we do not know if clk-2 ts alleles are completely inactive when shifted to the restrictive temperature during early embryonic cell cycle progression . Indeed , as is the case for the germ cell cycle arrest phenotype , a complete inhibition of clk-2 might result into earlier or more severe defects during embryonic cell divisions potentially resembling the clk-2 ( tm1528 ) germ cell cycle arrest phenotype . We extensively tried RNAi to completely inhibit clk-2 during embryogenesis using both RNAi injection and feeding procedures but never found a phenotype stronger than the phenotype of either clk-2 ts allele propagated at 25°C ( data not shown ) . clk-2 RNAi injections did not result in any phenotype [5] , and only the RNAi feeding construct introduced by the Nilsen laboratory worked for RNAi feeding . Only , when we analyzed clk-2 mutants kept at 25 . 5°C combined with clk-2 ( RNAi ) we observed more severe defects as seen in clk-2 ( qm37 ) and clk-2 ( mn159 ) mutants at the restrictive temperature . Under these conditions we observe a further delay of cell cycle progression ( particularly in the P lineage ) as compared to clk-2 ts mutants kept at the restrictive temperature ( Figure S5 ) . This delay appears as atl-1 independent . ATL-1 dependence was , however , difficult if not almost impossible to study due to severely abnormal cell divisions ( data not shown ) , that often resulted in cell divisions where only one daughter cell received an intact nucleus . Even without atl-1 ( RNAi ) treatment , nuclei often appeared as disorganized and at times fragmented under DIC optics ( Figure S5 ) , but we never observed uniform defects starting from the very first cell cycle , further complicating a detailed analysis ( data not shown ) . Obviously , these findings will raise the question as to how CLK-2 might affect early embryonic cell divisions , which will be the subject of further studies . These studies will however , require new clk-2 alleles as we currently can not rule out the possibility of off target effects associated with clk-2 RNAi that might unspecifically enhance clk-2 mutant defects . At present , we can only speculate if the developmental , cell cycle related and DNA damage response pathway defects associated with clk-2 mutations are due to a single molecular defect . We , indeed , favour an alternative model according to which CLK-2 affects multiple molecular processes . Our analysis which is based on an allelic series of clk-2 mutants with increasing strength clearly indentifies distinct functions associated with CLK-2 during embryonic and germ cell cycle progression as well as during embryonic development . It was recently shown that TEL2/CLK-2 belongs to the ARM repeat superfamily of structurally related proteins [47] ( Alexander Schleiffer , personal communication ) . Tandem ARM repeats fold together into a superhelical fold to form a surface for protein–protein interactions ( for review see , [51] , [52] ) . ARM repeat proteins are structurally related to proteins containing tandem HEAT motifs [51] . The demonstrated interactions between Tel2/CLK-2 and the HEAT repeat containing PIKKs suggests that TEL2/CLK-2 might act as an adaptor protein that impinges on multiple signalling pathways besides PIKKs through ARM/HEAT domain mediated protein-protein interactions . Our dissection of CLK-2 phenotypes in C . elegans is likely to stimulate future studies in mammalian cells addressing developmental and cell cycle-related functions of CLK-2/TEL-2 .
C . elegans strains were maintained at 20°C unless otherwise stated as described [53] . The following strains were used: clk-2 ( mn159 ) [5] , clk-2 ( qm37 ) [54] , cep-1 ( lg12501 ) [55] , gld-2 ( q497 ) gld-1 ( q485 ) ( gift of Tim Schedl ) , div-1 ( or148 ) [49] , clk-2 ( tm1528 ) was generated and kindly provided by Shohei Mitani . The clk-2 ( tm1528 ) deletion strain was backcrossed 5 times to reduce background mutations and balanced with hT2 [bli-4 ( e937 ) q418] by crossing to JK2689 [pop-1 ( q4645 ) dpy-5 ( e61 ) /hT2 [bli-4 ( e937 ) q418] to generate TG56 clk-2 ( tm1528 ) /hT2 [bli-4 ( e937 ) q418] . Further strains used were TG58 cep-1 ( lg12501 ) ; clk-2 ( qm37 ) , TG57 cep-1 ( lg12501 ) ; clk-2 ( mn159 ) , TG59 cep-1 ( lg12501 ) ; div-1 ( or148 ) , TG60 gld-2 ( q497 ) gld-1 ( q485 ) /hT2 [bli-4 ( e937 ) q418]; clk-2 ( tm1528 ) /hT2 [bli-4 ( e937 ) q418] . RNAi was performed by using the feeding procedure [56] . RNAi-expressing bacteria were seeded on NGM agar plates containing 3 mM IPTG and 50 µg/ml ampicillin , and worms were added as L4 larvae the following day . Animals were fed with bacteria carrying an empty L4440 feeding vector [57] or atl-1 , chk-1 [15] and mdf-1 feeding ( MRC geneservice ) constructs . Phenotypes were observed in F1 animals . F1 animals in the L4 stage were placed onto RNAi plates . F2 embryos were analysed after 24 h of incubation , and F1 animals were analysed after 48 h to observe germline phenotypes . Worms at the indicated time post-L1 were stained by DAPI using the following procedure . Animals were transferred to 100 µl M9 buffer and washed 3× with M9 buffer and resuspended in 1 ml 96% ethanol containing DAPI ( 200 ng/ml ) for 1 h and rehydrated with 1 ml M9 buffer for 1 h . Worms were transferred into 3 µl of mounting solution ( 90% glycerol , 20 mM Tris pH 8 . 0 , 1 mg/ml p-phenylenediamine ) and mounted on slides . Germ cells were identified by nuclear morphology according to DAPI staining . For the antibody staining , one day post-L4 adult gonads ( for clk-2 ( tm1528 ) 48 h post L4 ) were dissected in EBT ( 25 mM HEPES pH 7 . 4 , 0 . 118 M NaCl , 48 mM KCl , 2 mM CaCl2 , 2 mM MgCl2 , 0 . 1% Tween 20 ) on a slide coated with poly-lysine ( Sigma ) and freeze-cracked . The slides were transferred to −20°C cold methanol , for 5 minutes and washed three times in PBS for 10 minutes at RT . Slides were blocked for 30 minutes in 0 . 5% BSA in PBST ( PBS , 0 . 05% Triton-X100 ) and incubated overnight at 4°C with the primary antibody ( 1/1000 in 3% BSA in PBST ) . The next day , the gonads were washed three times in PBST each for 5 minutes at RT and incubated with the secondary antibody for 1 hour at room temperature . Gonads were washed three times in PBST each for 10 minutes and mounted with 5 µl mounting solution containing 0 . 5 µg/ml DAPI . Antibodies were used at the following dilutions: anti-α-tubulin antibody DM1A ( Sigma ) was used at 1/200 , anti-γ-tubulin 1/5000 ( gift of Carrie Cowen , IMP Vienna ) , anti PH3 1/400 ( Upstate ) , anti RAD-51 1/200 [25] , anti-Cdk1 1/100 ( pTyr15 , Calbiochem ) . Secondary antibodies used were anti-rabbit cy3 and anti-mouse FITC ( 1/1000 , Jackson ) . Methods for 4D-microscopy were described in [13] . Modifications of the 4D-microscope system are described in [31] . Embryos were recorded at 25°C and stacks of 25 DIC-images , viewing different focal planes of the developing embryo , were taken every 35 sec . The 4D-recordings were analysed using the SIMI Biocell program ( SIMI Reality Motion Systems , Unterschleissheim , Germany; http://www . simi . com ) [31] , [13] . Cell cycle timing was determined by measuring the time between the two mitotic divisions ( completion of cytokinesis ) . Deltavision microscopy was used to examine germlines using either a 60× or a 100× , UPlanSApo objective ( Olympus; NA 1 . 40 ) , Soft-WoRx software ( Applied Precision ) , and a CoolSnap HQ ( Photometrics ) OCD camera . Three-dimensional datasets were computationally deconvolved , and regions of interest were projected onto one dimension . Protein samples were resolved by SDS-PAGE analysis and transferred to polyvinylidine difluoride membrane ( PVDF , Millipore ) . Membranes were blocked in 5% powdered milk , diluted in PBS Tween , then probed with primary antibody diluted in blocking solution for 3 hours . Primary antibodies were anti-CEP-1 ( 1/100 [55] ) and anti-CLK-2 ( 1/1000 gift of S . Boulton ) . Antibody binding was detected using anti-rabbit or anti-goat IgG coupled to horse radish peroxidase ( Jackson ) and proteins were visualized using ECL ( Amersham ) and autoradiography . | PI3K-related protein kinases ( PIKKs ) ATM and ATR are essential upstream components of DNA damage signalling pathways , while TOR-1 acts as a nutrient sensor . CLK-2/TEL2 is a conserved gene initially implicated in budding yeast telomere length regulation and uncovered in the same genetic screen as the yeast TEL1 ATM like kinase . CLK-2/TEL2 was first implicated in DNA damage response signalling by C . elegans genetics , a function confirmed in yeast and human cells . In addition , CLK-2/TEL2 is essential for cellular and organismal survival from yeasts to vertebrates , but the essential phenotypes were not defined . A direct interaction between CLK-2/TEL2 and all PI3K-related protein kinases and the reduction of PIKK protein levels upon CLK-2/TEL2 depletion lead to the widely discussed notion that CLK-2/TEL2 mutants might phenocopy PIKK depletion phenotypes . We take advantage of embryonic lineage analysis and germline cytology to dissect developmental and cell cycle related functions of CLK-2 . CLK-2 depletion does not phenocopy PIKK kinase depletion . We rather link CLK-2 to multiple developmental and cell cycle related processes and show that CLK-2 and ATR have antagonising functions during early C . elegans embryonic development . Furthermore , we implicate CLK-2 in a distinct cell lineage decision and show that its depletion leads to a novel germline cell cycle arrest phenotype . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] | [
"cell",
"biology/cell",
"growth",
"and",
"division",
"cell",
"biology",
"genetics",
"and",
"genomics/gene",
"function",
"genetics",
"and",
"genomics/animal",
"genetics"
] | 2009 | Functional Dissection of Caenorhabditis elegans CLK-2/TEL2 Cell Cycle Defects during Embryogenesis and Germline Development |
Buruli ulcer ( BU ) is a skin disease caused by Mycobacterium ulcerans . Its exact mode of transmission is not known . Previous studies have identified demographic , socio-economic , health and hygiene as well as environment related risk factors . We investigated whether the same factors pertain in Suhum-Kraboa-Coaltar ( SKC ) and Akuapem South ( AS ) Districts in Ghana which previously were not endemic for BU . We conducted a case control study . A case of BU was defined as any person aged 2 years or more who resided in study area ( SKC or AS District ) diagnosed according to the WHO clinical case definition for BU and matched with age- ( +/−5 years ) , gender- , and community controls . A structured questionnaire on host , demographic , environmental , and behavioural factors was administered to participants . A total of 113 cases and 113 community controls were interviewed . Multivariate conditional logistic regression analysis identified presence of wetland in the neighborhood ( OR = 3 . 9 , 95% CI = 1 . 9–8 . 2 ) , insect bites in water/mud ( OR = 5 . 7 , 95% CI = 2 . 5–13 . 1 ) , use of adhesive when injured ( OR = 2 . 7 , 95% CI = 1 . 1–6 . 8 ) , and washing in the Densu river ( OR = 2 . 3 , 95% CI = 1 . 1–4 . 96 ) as risk factors associated with BU . Rubbing an injured area with alcohol ( OR = 0 . 21 , 95% CI = 0 . 008–0 . 57 ) and wearing long sleeves for farming ( OR = 0 . 29 , 95% CI = 0 . 14–0 . 62 ) showed protection against BU . This study identified the presence of wetland , insect bites in water , use of adhesive when injured , and washing in the river as risk factors for BU; and covering limbs during farming as well as use of alcohol after insect bites as protective factors against BU in Ghana . Until paths of transmission are unraveled , control strategies in BU endemic areas should focus on these known risk factors .
Buruli ulcer ( BU ) is a chronic debilitating skin disease caused by Mycobacterium ulcerans [1] , [2] . BU depicts the third and second most common mycobacterial disease , globally and in Ghana , respectively [3] , [4] . Currently , BU has been reported in over 30 countries in four continents [1] , [2] , [5] , [6] but West Africa is the region most affected [1] , [6] . The first case of BU in Ghana was reported in 1971 by Barley [7] , [8] , and ever since over 426 communities have reported cases . These communities are in the Ashanti , Brong Ahafo , Eastern , Greater Accra and Western regions . Amofah et al found the highest prevalence rate of 87 . 7/100 , 000 in the Ga West District [9] . Jacobsen and Padgett systematically reviewed extensive epidemiological studies done to identify risk factors associated with M . ulcerans throughout the world . The commonly reported risk factors associated with BU were slow flowing or stagnant water [4] , [10]–[13] , wading [14] , [15] or washing clothes in swampy areas of slow flowing waters [16] , and the use of short clothes during farming [15] , [16] . Merritt et . al . reported similar risk factors in their systematic review on ecology and risk factors for transmission of BU [6] . Other risk factors reported were close proximity to human disturbed aquatic habitats [6] , the use of unprotected water from swamps [17] and rivers [4] , [7] , and agricultural land use [18] . Reduced risk for BU , however was associated with the use of protected water sources in some settings [14] , [17] as well as hygienic practices such as use of soap for bathing , use of alcohol to clean wounds , or injured sites and proper wound care [4] , [14] , [15] . Researchers in Amansie West District of Ghana demonstrated spatial relationship between BU prevalence and the immunosuppressant arsenic [13] . With regard to the role of insect bites in the transmission of M . ulcerans , water bug ( aquatic Hemipterans ) species have been particularly addressed [19] , [20] , [21] , [22] , [23] . Series of studies demonstrated mosquitoes and water bugs to carry M . ulcerans in endemic areas [24] , [25] . Australian studies showed association of mosquito related risk factors with BU [26] , [27] , and experimental infection of mice bitten by infected water bugs in laboratory provided evidence to support their involvement [21] , [28] . The argument for mosquitoes as vectors gained more ground when the use of bed nets was found to reduce the risk of BU [4] , [15] , [29] . Children aged less than fifteen years are overrepresented compared to adults albeit any age can be affected [6] , [7] , [30] , [31] . Even though such risk factors have been identified , the exact mechanism by which humans contract BU in or near aquatic habitats is still not known . It has been hypothesized that M . ulcerans is transmitted through skin abrasions or skin injuries after contact with water , vegetation , or soil which still remains a hypothesis [18] . Without knowing the exact mode of transmission , the only recommendations to effectively prevent and control BU should be based on the currently known risk factors . SKC and AS Districts of the Eastern Region in Ghana have been recently identified as BU endemic but data on the prevailing risk factors was not yet available . Here , we conducted a case control study to identify the risk factors for BU in these previously non-endemic districts .
A case-control study was designed in two health districts , SKC and AS of the Eastern Region . The cases were identified through active community case search by trained Community Based Volunteers ( CBVs ) ( Figure 1 ) . Information was provided to all members of the various communities and subsequently individually to the participants . Enrollment into the study was voluntary . All adult subjects provided written informed consent and a parent or guardian of any child participant provided written informed consent on their behalf . Ethical clearance was obtained from the NMIMR Institutional Review Board and the Ghana Health Service Ethical Committee . The approval was renewed yearly during the period of the study . We used the power calculation tool ( Epi Info software version 3 . 5 . 1 ) to determine the sample size . We set alpha to 0 . 05 and power equal to 80% . The districts reported 40% use of unprotected water . The minimum of the odds ratio ( OR ) for the association between cases and controls was set at 2 . 25 . We obtained a sample size of 214 participants , made up of 107 cases and 107 controls . BU active community case search was conducted by trained CBVs from May 2010 to December 2011 . The research team was introduced to the head of the community , opinion leaders and solicited their cooperation on the research being carried out . Sensibilization was given to the community members through town cries “gongon beater” and by information using posters and pictures of BU prior to physical examination . Research assistants administered standardized questionnaires that covered issues on demography ( age , gender , place of residence , marital status , occupation , and educational status ) , and behavioral activities ( swimming , wading , fishing , wearing of protective clothing and personal hygiene ) . In addition , environmental issues ( nearby presence of wetland/swamp , vegetations , cocoa or coffee plantations , sources of drinking water , type of houses , sharing of living space with animals/pets and other peculiar characteristics of the locations were also assessed . All questions were closed-ended and the questionnaires were verbally administered in English or the local language , Twi . Bacille Calmette-Guérin ( BCG ) vaccination was assessed by observing for the presence of the scar on the left shoulder around the deltoid region as vaccination cards were difficult to assess and in some cases missing . Wound swabs from ulcers and fine needle aspirates from nodules were used for laboratory confirmation . BU was the dependent variable and demographic , host related , environmental and behavioural factors as the independent variables . Significance level was set at a p-value less than 0 . 05 . Univariate analysis was done using conditional logistic regression to calculate odds-ratios ( OR ) and 95% confidence intervals ( 95% CI ) to explore the association between the exposure variables and BU . All variables obtained from the univariate analysis with p-values ≤0 . 1 were retained for the multivariate model . The variables in the final model were retained after a step-by-step backward elimination using multiple conditional logistic regression .
A total of 141 probable BU patients were enrolled , from which 113 ( 80 . 1% ) were confirmed PCR positive . Among those , 66 ( 58 . 4% ) were also positive for Ziehl-Neelsen stain . The median age of the confirmed cases was 28 years ( ranging from 2 to 102 years ) . The commonest age group affected was above 24 years with 54 . 9% ( 62/113 ) . Among the case patients 50 . 4% ( 57/113 ) were female and 49 . 6% ( 56/113 ) were male . In addition to the various BU active lesions , contracture deformities were observed in twelve of the cases with active lesions , extensive scar due to BU in five of the cases and one patient had had amputation of the right little toe . Ethnic group distribution of the parents of the participants were Akan , Ewe and Ga Adangme . For parental ethnic groups 35 . 4% ( 40/113 ) of fathers of case patients were of Akan ethnic group , 50 . 4% ( 57/113 ) were Ewe and 14 . 2% ( 16/113 ) were Ga Adangme ( Table 1 ) . Wading in the Densu river was more frequent among the case patients than the community controls and was significantly associated with BU ( OR = 3 . 5 , 95% CI = 2 . 0–6 . 1 ) . However , wading in other rivers or streams , fetching of water and fishing in Densu river were not significantly associated with BU . Taking a bath with water taken from an open borehole was more frequent among case patients than community controls ( OR = 3 . 5 , 95% CI = 1 . 4–8 . 5 ) . With no farming as a reference point , there was significant association between farming with long sleeves and BU ( OR = 0 . 31 , 95% CI = 0 . 16–0 . 57 ) , long pants ( OR = 0 . 25 , 95% CI = 0 . 13–0 . 5 ) but not significantly associated with use of short sleeves and short pants when farming ( Table 3 ) . Rubbing the area with alcohol after an insect bite ( 0 . 21; 95% CI = 0 . 008–0 . 57 ) and farming with long sleeve clothes ( 0 . 29; 95% CI 0 . 14–0 . 62 ) were found to be protective factors . Insect bite in water/mud ( OR = 5 . 7 , 95% CI = 2 . 5–13 . 1 ) , presence of wetland ( OR = 3 . 9 , 95% CI = 1 . 9–8 . 2 ) , use of adhesive bandage ( OR = 2 . 7 , 95% CI = 1 . 1–6 . 8 ) , wading in Densu river ( OR = 2 . 3 , 95% CI = 1 . 1–4 . 96 ) and house wall built with mud ( OR = 2 . 6 , 95% CI = 1 . 1–5 . 9 ) were risk factors associated with BU ( Table 4 ) .
This study identified activities that showed statistically significant association with BU in SKC and AS Districts of the Eastern region of Ghana , an area recently identified as being endemic for BU . Farming with long sleeve clothes and rubbing an insect bite area with alcohol were associated with decreased risk of contracting BU . On the other hand , presence of wetland , insect bites in water/mud , washing in the Densu river , use of adhesive bandage and house walls built with mud were identified as risk factors for BU . Without doubt , all limitations associated with the case control study approach apply to this investigation . Most of the case patients have been living with the disease for more than two years , hence prevalent cases rather than incident cases were recruited . For a chronic and rare disease like BU , association of disease persistence may be confounded with disease development . Also , recall bias remained a major limitation to this study , both from case patients and respondent parents on behalf of their wards . However , the interviewers were trained to ensure that appropriate responses were elicited from the respondents so as to minimize any form of bias or confounding effects to the findings . This study comes sequent to several epidemiological studies identifying risk factors for transmission of BU [4] , [7] , [13] , [14] , [15] , [25] , [29] , [32] , [33] , [34] , and our findings validate in the Eastern Region of Ghana what has been reported in other countries . Ulcerative forms of disease presentation constituted 84% ( 95/113 ) of all cases . This implies that most of the case patients presented or were diagnosed late , probably due to factors such as transportation costs , feeding costs , and productivity loss [3] , [35] , [36] . This may be the underlying reason for the high median age of the participants in the study . It was found that 67 . 9% ( 76/113 ) of the case patients had lesions on their lower limbs [15] , [37]–[39] albeit with no preference to either side of the body . This is in contrast to an earlier study done in the Ashanti Region of Ghana reporting more frequent affection of the left leg [24] , a finding which could also not be confirmed by other studies [37] , [40] . Concerning earlier findings of predisposition for or genetic link to BU [33] the present result show ( albeit not significant in the multivariate model ) that BU was less common in the Akan ethnic group . No significant relationship was found between anamnesis of a past tuberculosis [15] nor to a protective role of BCG vaccination to BU , as indicated by previous reports [4] , [15] , [26] , [30] , [41] , [42] . Case patients reported more frequently insect bites in water or wading in mud than the community controls did , which was evident as statistically significant in other studies [15] , [26] . This finding tends to support the hypothesis that M . ulcerans can only enter the body through broken skin due to either insects bites or abrasions . Likewise , an appropriate initial treatment upon injury like rubbing the area with alcohol seems to offer protection against development of BU . Surprisingly , the use of adhesive bandage when hurt increased the odds of contracting BU , probably owing to the fact that often adhesive bandages were already being used by other persons and thus contaminated . In fact , most such bandages looked old and dirty . Wading , swimming , and fishing in the Densu river were not identified as risk factors for BU . Swimming was not widely practiced in the study area [7] although a study in Cote d'Ivoire found such an association [16] . The type of fishing undertaken in the Eastern Region of Ghana differs from habits in many areas that did identify correlations to fishing activities [4] , [15] , [16] . Here , commonly either lines with hooks or small nets are being placed at the bank of the river hence resulting in little or no contact to water . The present study confirms , however , findings of other studies [4] , [15] , [16] that arming with long sleeves and long pants protects against BU . Long clothes may protect from small injuries or insect bites as possible means of entry for M . ulcerans . In line with previous studies , the use of soap for washing was found to be associated with a decreased risk of M . ulcerans infection [4] , [14] . In order to approach the role of mosquitoes in the transmission of BU , we used the protection of bed nets as a proxy to assess association to contracting BU . In accordance with Raghunathan's finding [4] , this study showed no evidence for protective effects of bed net usage . Since other studies showed the contrary [14] , [15] , [26] we reason that in malaria endemic countries , the role of mosquitoes in the transmission of BU may be under investigated . Likewise , and also in contrast to earlier reports [4] , [15] , the present study showed no evidence of association between the use of mosquito coils and BU . In this newly identified BU endemic area of the SKC and AS Districts in the Eastern Region of Ghana , our study identified as risk factors the presence of wetlands , insect bites in water , use of adhesive when injured and washing in the Densu river . In contrast , covering limbs during farming and use of alcohol after insect bites were found to be protective factors for BU . Until the mode of transmission is completely unraveled , provision of information in public health measures and steadily raising awareness of these risk factors are important means to both prevent and control BU . | Mycobacterium ulcerans is the causative agent of Buruli ulcer ( BU ) which affects the skin , can lead to extensive ulceration , and often results in disabilities . The exact mode of transmission of the disease is still unknown . Previous studies have identified demographic , socio-economic , health and hygiene , as well as environment , related risk factors for BU . This case-control study was done to ascertain the risk factors in a study area in Ghana which was previously non-endemic for BU . The study involved 226 participants , of which 50% were BU confirmed cases and 50% age- , gender- , and community matched controls ( persons who had no signs or symptoms of active or inactive BU ) . This study found presence of wetland , insect bites in water , use of adhesive when injured , and washing in the Densu river as risk factors associated with BU . These factors were similar to previous studies and hence should be used in the implementation of national BU control strategies . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences"
] | 2014 | Risk Factors for Buruli Ulcer in Ghana—A Case Control Study in the Suhum-Kraboa-Coaltar and Akuapem South Districts of the Eastern Region |
DNA replication programs have been studied extensively in yeast and animal systems , where they have been shown to correlate with gene expression and certain epigenetic modifications . Despite the conservation of core DNA replication proteins , little is known about replication programs in plants . We used flow cytometry and tiling microarrays to profile DNA replication of Arabidopsis thaliana chromosome 4 ( chr4 ) during early , mid , and late S phase . Replication profiles for early and mid S phase were similar and encompassed the majority of the euchromatin . Late S phase exhibited a distinctly different profile that includes the remaining euchromatin and essentially all of the heterochromatin . Termination zones were consistent between experiments , allowing us to define 163 putative replicons on chr4 that clustered into larger domains of predominately early or late replication . Early-replicating sequences , especially the initiation zones of early replicons , displayed a pattern of epigenetic modifications specifying an open chromatin conformation . Late replicons , and the termination zones of early replicons , showed an opposite pattern . Histone H3 acetylated on lysine 56 ( H3K56ac ) was enriched in early replicons , as well as the initiation zones of both early and late replicons . H3K56ac was also associated with expressed genes , but this effect was local whereas replication time correlated with H3K56ac over broad regions . The similarity of the replication profiles for early and mid S phase cells indicates that replication origin activation in euchromatin is stochastic . Replicon organization in Arabidopsis is strongly influenced by epigenetic modifications to histones and DNA . The domain organization of Arabidopsis is more similar to that in Drosophila than that in mammals , which may reflect genome size and complexity . The distinct patterns of association of H3K56ac with gene expression and early replication provide evidence that H3K56ac may be associated with initiation zones and replication origins .
DNA replication is a fundamental process required for the growth and development of all eukaryotes . This process is regulated both spatially and temporally so that all DNA sequences are replicated exactly once during S phase , insuring that each daughter cell receives a complete copy of the genome . DNA replication initiates from discrete locations on chromosomes known as replication origins ( origins ) where proteins required for DNA synthesis are recruited by the origin recognition complex ( ORC ) . Once initiated , DNA replication proceeds by elongation to regions where opposing replication forks converge ( termination zones ) . This organization of DNA sequences into regions of initiation , elongation and termination define a replicon – a segment of DNA replicated as a unit by replication forks originating from a single origin [1]–[5] . The time of replication for any particular DNA sequence within a replicon is determined by three factors: its proximity to an origin , the efficiency of initiation at that origin , and the rate of DNA elongation in that region . The pattern of DNA replication has been determined for multiple eukaryotic genomes ranging from the compact genome of budding yeast to the moderately sized genome of Drosophila melanogaster and the large human and mouse genomes [6]–[14] . In budding yeast , DNA sequences acting as origins have a conserved consensus motif , and origin activation appears to follow a strict temporal program [6] . However , recent single molecule studies of DNA replication in yeast [15] , [16] suggest that the temporal program likely represents the average replication program for a population of cells , with considerable variation in the order of origin activation in individual cells [17]–[20] . In higher eukaryotes , no consensus sequence for origin DNA has been identified , and some known origins are organized as broad initiation zones containing multiple potential origins [2]–[4] . It is unclear whether origin activation follows a temporal sequence in higher eukaryotes , but origin activation in Drosophila is most prevalent in early and late S phase , suggesting some degree of temporal regulation [14] . In mammals , clusters of replicons frequently display coordinate origin activation and are organized into larger replication domains [1] , [5] , [21] . The organization of replication domains appears to be cell type specific , as differentiation of embryonic stem cell lines to neural precursor cells resulted in the widespread reorganization of replication domains [13] . Differences in replication patterns between cell types have been linked to changes in gene expression and epigenetic modifications [13] , [14] . The relationship between gene expression and replication time has been examined in yeast , Drosophila , mouse and human cells . In budding yeast , there is little correlation between replication time and gene expression [6] . In higher eukaryotes with more complex genomes , there is a positive correlation between early replication and gene expression , and this correlation is strongest when integrated over large chromosomal domains [7] , [8] , [10]–[14] , [22] , [23] . The fact that an open chromatin conformation is necessary but not sufficient for both DNA replication and gene expression may underlie the correlation between these processes [2]–[4] , [11] , [21] , [24] . In general , euchromatin replicates early in S phase and heterochromatin replicates late , although specific types of heterochromatin replicate in early S phase in yeast [6] , [8] , [21] , [25] , [26] . Chromatin is subject to a plethora of epigenetic modifications including histone methylation , histone acetylation and DNA cytosine methylation ( 5mC ) . The combinatorial effect of these modifications , as well as the association of other chromatin-binding proteins , determines whether DNA adopts a heterochromatic or euchromatic conformation [27]–[29] . Epigenetic modifications associated with heterochromatin and characteristic of silenced genes and transposable elements include tri- and dimethylation of histone H3 lysine 9 ( H3K9me ) , hypoacetylation of histones , and abundant 5mC [27]–[34] . There are conflicting reports for the correlation between heterochromatic marks and late replication , which is surprising given the tight relationship between late replication and heterochromatin [12] , [13] , [35] . Modifications associated with euchromatin and active or potentially active genes include tri- , di- and monomethylation of histone H3 lysine 4 ( H3K4me ) , hyperacetylation of histones , and 5mC localized to gene coding sequences [27]–[30] , [32]–[34] , [36]–[38] . Several replication timing studies showed a positive correlation of early replication with H3K4me [12] , [13] , [35] , which may be indirect because H3K4me is associated almost exclusively with genes and gene-rich regions which tend to replicate early [7] , [8] , [11] . Several lines of evidence suggest that the link between histone acetylation and replication time is more direct . Hyperacetylaton of histone H3 on lysines 9 and 14 ( H3K9/14ac ) associates with origins in human cells [39] . Hyperacetylation of histone H3 lysine 56 ( H3K56ac ) associates with early firing origins in budding yeast [40] . Hyperacetylation of histone H4 lysine 16 ( H4K16ac ) associates with early replicating regions in Drosophila cells [14] . In addition , late-firing origins in budding yeast are regulated by a histone deacetylase complex [41] . These and other experiments suggest that histone acetylation may be the best epigenetic determinant of replication time [24] , [42]–[44] . Very little is known about the regulation of DNA replication in plants [45] . The core proteins required for DNA replication are conserved between yeast , plants and animals [46] , [47] . The replication machinery of plants is more similar to animals than yeast , but many of the genes encoding these proteins have multiple homologs in Arabidopsis thaliana suggesting that functional diversification has occurred [47] . DNA fiber autoradiography studies revealed that Arabidopsis possesses two families of replicons , one initiating replication early and the other later in S phase [48] . These likely correspond to euchromatic and heterochromatic replicons because , like most eukaryotes , plants replicate heterochromatin later than euchromatin [25] . In contrast , knowledge of epigenetic modifications in Arabidopsis has kept pace with other systems , and with few exceptions , these modifications are functionally conserved between plants and animals [28] , [29] , [49] . The relationship between epigenetic modifications and DNA replication in plants is virtually unexplored . However , DNA replication is required to maintain the repressed state of a negative regulator of flowering in Arabidopsis [50] , suggesting that the interplay of these processes is crucial for plant growth and development . Similar to the replication machinery , the genes encoding DNA and histone modifying enzymes often have multiple homologs in plants [51]–[53] . Arabidopsis with its small , well-characterized genome is an excellent model system for examining the global relationship between DNA replication and chromatin state in higher eukaryotes . The genome of Arabidopsis is gene-dense in comparison to mammalian genomes , with roughly the same number of genes encoded by a genome one-twentieth the size [54]–[56] . The genome size of Drosophila is similar but encodes half the number of genes [57] . This characteristic of Arabidopsis may provide insight into the influence of gene density on DNA replication . In addition , analysis of Arabidopsis DNA replication has the potential to uncover features that are unique to plants . The diversification of genes encoding replication-associated proteins and chromatin modifiers suggests that plants may have developed unique mechanisms to regulate DNA replication and to establish and maintain chromatin states . These mechanisms may be related to developmental pathways that are common in plants but rare in other systems . For example , endoreduplication plays a prominent role in plant development and totipotency of plant cells is not limited to germline or embryonic cells . We used a combination of fluoresence-activated cell sorting ( FACS ) and genomic tiling arrays to profile DNA replication of Arabidopsis chr4 in early , mid and late S phase cells . We investigated the relationship between DNA replication , gene expression and chromatin state in analyses of our data and the extensive genomic data available for Arabidopsis chr4 .
We used an established Arabidopsis Col-0 suspension cell line for the analysis of replication time and optimized the culture conditions to provide ample nuclei from replicating cells for fractionation by FACS ( Figure S1 and Text S1 ) . This cell line was also used in recent studies that examined the effects of cell culture on specific epigenetic modifications [34] . We first characterized the relationship between DNA content and replication in this cell line by monitoring the incorporation of the nucleotide analog bromodeoxyuridine ( BrdU ) . An asynchronous population of cells was labeled with BrdU for 1 hour and fixed . Nuclei were isolated , stained with propidium iodide , labeled with a fluorescent anti-BrdU antibody , and analyzed by FACS for DNA content and BrdU incorporation . Nuclei in S phase that incorporated BrdU appeared as a distinct “arc” above the population of cells in G1 and G2/M ( Figure 1A ) . Surprisingly , almost 30% of the S phase nuclei fractionated above the G1 peak , and we designated this population early S phase ( Figure 1A and Table S1 ) . Similarly , we designated the 50% of the S phase nuclei that fractionated above the G2/M peak as late S phase . The remaining 20% of S phase nuclei between the G1 and G2/M peaks were designated mid S phase . We estimated the DNA content of the early , mid and late S phase populations at 1 . 16 , 1 . 49 and 1 . 95C , respectively ( Figure 1B ) . This distribution of S phase nuclei and DNA content indicated that to get a complete picture of DNA replication during S phase we needed to analyze DNA replication in nuclei that co-sorted with G1 ( early S phase ) and G2/M ( late S phase ) peaks . We profiled DNA replication independently in early , mid and late S phase . We could not sort the early , mid and late S phase nuclei based on BrdU content because visualization of the BrdU degrades DNA . Instead , nuclei were sorted based on DNA content , and BrdU-labeled DNA was separated by immunoprecipitation ( Figure 1C ) . Nuclei in the early S/G1 , mid S and late S/G2/M sorts contained different fractions of nuclei in S phase , with the early S/G1 , mid S and late S/G2/M sorts containing 4 . 2 , 42 . 3 and 18 . 3% BrdU-positive nuclei , respectively ( Figure 1A and Table S1 ) . Because of these differences , it was necessary to account for cross contamination associated with sorting ( Figure S2A ) , especially contamination of mid S phase nuclei into the early S/G1 sort ( Table S2 and Figure S2B ) . When corrected for the percentage of nuclei in S phase , we determined that the early , mid and late S phase purity was 69 , 94 and 85% respectively ( Table S2 ) . In the worst case , 28% of S phase nuclei in the early S/G1 sort were actually in mid S phase ( Table S2 and Figure S2B ) . However , this contaminating population had a DNA content from the lower tail of the mid S phase distribution ( Figure S2B ) . BrdU-labeled DNA from early , mid or late S phase nuclei was hybridized separately to a tiling microarray that covers 99% of the sequenced regions of chr4 of Arabidopsis thaliana with 22 , 761 PCR-generated probes averaging 1 kb in length [33] . This array was used previously to profile specific epigenetic modifications in this cell line [34] . Microarray results were confirmed by qPCR analysis of 14 selected regions ( Tables S3 and S4 , Figure S3 and Text S1 ) . Figure 2 shows a schematic representation of chr4 including plots for gene and transposable element ( TE ) coverage ( Figure 2A ) and GC content ( Figure 2B ) . Chr4 is unusual in that it has three regions of constitutive heterochromatin – the nucleolar organizing region ( NOR ) at the end of the short arm ( not shown ) , a 700 kb heterochromatic knob centered at 2 Mb , and 2 . 5 Mb of pericentromeric heterochromatin centered at 4 Mb ( Figure 2C ) [58] , [59] . These heterochromatic regions were used as boundaries to subdivide chr4 into six regions for subsequent analyses – the distal short arm , the heterochromatic knob , the proximal short arm , the pericentromere , the proximal long arm and the distal long arm ( Figure 2C ) . The boundaries of most of these regions are evident from gene and transposable element ( TE ) coverage and to some extent from the GC content profile ( Figure 2A and 2B ) . The boundary between the proximal and distal long arms is less evident and was chosen based on the replication time results presented below . The replication profiles were generated from the microarray data by applying a loess algorithm in a 150-kb window to smooth the probe-level data ( Figure 2D ) . The early and late profiles display remarkable complementarity ( R = −0 . 83 ) , i . e . regions of chr4 enriched for BrdU in early S phase cells are depleted in late S phase cells . Early replication is most prevalent in the distal long arm , a euchromatic region rich in genes with few TEs . Late replication predominates in the heterochromatic knob and pericentromere of chr4 , but regions of late replication are also dispersed in other parts of chr4 , most notably the proximal long and short arms . The replication profiles for early and mid S phase cells are surprisingly similar ( R = 0 . 87 ) ( Figure 2D ) . The most evident difference is a broadening and merging of early replicating regions in the mid S phase profile . The DNA replicating in mid S phase represents nearly the same population of sequences as that replicating in early S phase even though FACS analysis demonstrates that the early and mid S phase nuclei have notably different DNA content ( Figure 1 , Figure S2 and Tables S1 and S2 ) . Like early S phase , the mid S phase profile is distinct from the late profile ( R = −0 . 85 ) . The similarity of the early and mid S phase profiles is not consistent with a fixed order of origin activation and , instead , suggests that origin activation in early and mid S phase is stochastic . Together , the early , mid and late S phase profiles suggest that DNA replication in Arabidopsis cells is biphasic , a result consistent with a previous report that Arabidopsis DNA replication takes place in two distinct stages [48] . To facilitate further analyses , we performed a two-step segmentation of the early , mid and late S phase profiles to assign a replication time for each microarray probe . Figure 3 illustrates this process for two chr4 regions representative of early and late replicating regions . In the first step , we identified contiguous segments of probes showing coordinate replication times ( log2 ratio >0 ) within each smoothed profile , thereby defining segments of early , mid or late replicating DNA ( Figure 3C and 3D ) . In the next step , we reconciled the replication times between experiments by determining the regions of overlap between the early , mid and late segments ( Figure 3D ) . This analysis identified segments of DNA replicating only in early S phase ( E ) , in both early and mid S phase ( EM ) , only in mid S phase ( M ) , in both mid and late S phase ( ML ) , only in late S phase ( L ) , in early and late S phase ( EL ) , throughout S phase ( EML ) , and segments of indeterminate replication time ( I ) that did not show enrichment in any experiment ( Table S5 ) . The majority of chr4 replicates as either EM ( 37% ) or L ( 44% ) when segment length is taken into account ( Figure 4A ) . Only 4% of chr4 replicates exclusively in mid S phase ( M ) , while 6% replicates as ML and 6% replicates as EML . The positions of these segment types with respect to EM and L segments suggest that many of the M , ML and EML segments are regions of DNA elongation between EM segments or transition zones between early to late replication ( Figure 3 ) . In regions of predominately late replication , M , ML and EML segments are often located between larger flanking L segments ( Figure 3 ) , suggesting that they contain the DNA replication origins for the flanking regions . The EL segments comprise only 2% of chr4 and are enriched for repetitive sequences ( Table S6 ) . Thus , at least some EL segments are likely to be artifacts created by cross hybridization on the microarray . I segments , which comprise 2% of chr4 , also have an elevated repeat content ( Table S6 ) . Another possible explanation for EL and I segments is that replication time in these regions is driven by allele-specific gene expression and/or epigenetic modifications ( see below ) [12] , [60] . We also determined the replication times of the six chr4 regions defined in Figure 2 . The heterochromatic knob and pericentromere replicate almost exclusively as L segments while the gene-rich distal long arm replicates predominately as EM segments ( Figure 4A ) . The replication time of the distal short and proximal short regions is more complex , perhaps influenced by the flanking heterochromatic regions ( Figure 2 ) . The proximal long arm displays a surprising amount of late replication despite the fact that this region is not constitutive heterochromatin [59] , although it does have lower gene and higher TE content than the distal long region ( Figure 2 ) . Within a given replicon , the DNA closest to the origin will replicate earliest while the DNA located at termination zones , regions where opposing replication forks converge , will replicate latest . Replication time profiles have been used to identify both initiation and termination zones [6] , [7] , [14] . Initiation zones manifested as local maxima in the early and mid S phase profiles and as local minima in the late S phase profile ( Figure 3C ) . Conversely , termination zones manifested as local minima in the early and mid S phase profiles and as local maxima in the late S phase profile ( Figure 3C ) . We identified initiation and termination zones by computationally determining probes occurring at local maxima and minima in the loess smoothed profiles . We did not treat individual probes as initiation or termination zones and , instead , defined zones as 10 kb segments centered at the identified probes . Any zones that overlapped were then merged into a single zone . Replication time for each zone was determined from constituent probes . The number of initiation and termination zones was consistent between experiments ( Table S7 and Figure 3C ) . However , their positions were more consistent between the early and mid S phase profiles versus comparisons with the late S phase profiles , e . g . 80% of the initiation zones identified in the early S phase profiles are within 20 kb of an initiation zone identified in the mid S phase profiles while this figure dropped to 65% when comparing the early and late S phase initiation zones ( Table S8 ) . This difference is not unexpected given that initiation zones are more likely to replicate in early or mid S phase while termination zones are more likely to replicate in late S phase . We then examined the frequency of initiation and termination zones as a function of replication time . In Drosophila , initiation sites are more abundant in late replicating DNA than in early replicating DNA with very little initiation occurring in mid S phase [14] . In Arabidopsis , we found that the distribution of replication times for the initiation zones reflected the distribution of replication times for chr4 with initiation zones prominent only in EM ( 37% ) and L ( 42% ) segments ( compare Figure 4A and 4B ) . Thus , unlike Drosophila , initiation sites are no more abundant in late than in early replicating DNA . However , Arabidopsis appears similar to Drosophila [14] in that the majority of termination zones are located in L ( 61% ) rather than EM ( 19% ) segments ( Figure 4B ) . These results indicate that DNA replication in late S phase includes elongation from origins that have fired earlier in S phase as well as initiation and elongation from late firing origins . Higher eukaryotes do not possess replicons in the strictest sense of the term , but rather the concept of a “relaxed replicon” likely applies [2]–[4] . In this model , replication origins are not rigidly defined , and replicon boundaries can vary from cell to cell . We defined the boundaries of these relaxed replicons ( hereafter referred to as replicons ) using a subset of the termination zones . Where possible , we used termination zones that were identified in early , mid and late S phase cells . Where termination zones differed between experiments , we preferentially used termination zones enriched in late S phase cells or local minima from early or mid S phase cells for EM-replicating segments ( Figure 3C and Table S7 and Materials and Methods ) . In this way , we identified 164 termination zones that defined 163 putative replicons across chr4 with a median length of 107 kb . This replicon size is consistent with previous measurements of single replicons in Arabidopsis [45] , [48] , although we cannot exclude the possibility that at least some of these replicons are clusters of smaller replicons . The majority ( 154 ) have at least one putative initiation zone ( Figure 3C and Table S9 ) . This strategy worked well for the euchromatic regions of chr4 , particularly the distal long and distal short arms , where the predicted termination zones were consistent between early , mid and late S phase cells ( Figure 3C ) . There was less agreement between profiles for the other chr4 regions , and replicon boundaries in the late-replicating regions are defined primarily from the late S phase profiles ( Figure 3C and Table S7 ) . The assignment of a specific replication time to individual replicons is complex because a replicon can be comprised of DNA segments with replication times that cover the entirety of S phase . To simplify the analysis , we classified replicons based on the replication time of the probes comprising the greatest proportion of a replicon , e . g . a replicon comprised of 45% EM probes , 40% L probes and 15% M probes would be classified as EM . Figure 4C ( top panel ) shows a schematic representation of chr4 replicons with the replication times for the constituent probes . The complexity of replication time within replicons likely reflects several factors including time and efficiency of origin firing , the number of origins within initiation zones , and the rate of elongation by DNA polymerase in specific contexts [2]–[4] , [24] . In Drosophila , the interval between termination zones varies between early S phase and late S phase , with increased initiation in late S phase resulting in more closely spaced termination zones [14] . The size of the Arabidopsis replicons does not vary significantly between EM and L replicons ( Figure 4D ) . While M , ML and EML replicons are smaller than either EM or L replicons , the difference in size is not statistically significant ( Figure 4D ) . The similar size of EM and L replicons follows from the previous observation that initiation zones are no more abundant in late replicating regions than in earlier replicating regions ( Figure 4B ) . In mouse cells , replicons are organized into replication domains consisting of large clusters of replicons with similar replication times [13] , [22] . In Drosophila cells , clustering is less evident with replication profiles showing distinct peaks of early replication [14] . Arabidopsis appears more similar to Drosophila in this regard , but the 163 chr4 replicons could be organized into 41 replication domains based on their replication time ( Figure 4C , middle panel , and Table S10 ) . There are a few large replication domains , including a 4 . 5 Mb L domain ( coordinates 2 . 6–7 . 1 Mb ) that encompasses the entire pericentromere and portions of the proximal short and long arms , and a 2 . 3 Mb EM domain ( coordinates 16 . 2–18 . 5 Mb ) in the distal long arm ( Figure 4C , middle panel ) . However , the mean length of chr4 replication domains is 450 kb which is considerably smaller than the 1 Mb reported for mouse cells [13] . This difference in replicon organization may be related to genome size . The genome sizes of Arabidopsis and Drosophila are similar at 115 and 122 Mb , respectively [54] , [57] , while the mouse genome is estimated at 2500 Mb [56] . Replication time has been correlated with both genetic and epigenetic features in other model systems [7] , [8] , [11]–[14] . The replication profiles ( Figure 2 ) show that on the scale of the entire chromosome , EM replication is associated with euchromatic regions while L replication is associated with heterochromatic regions . To examine the relationships between replication time and both genetic and epigenetic features in more detail , we generated a database for computational analysis that incorporates our replication time data , the Arabidopsis TAIR 8 genome annotation [61] , and epigenetic information for the Arabidopsis cell line [34] . We performed our analyses both on the level of individual probes and within the context of replicons . To compare the genetic and epigenetic features of probes with different replication times , we partitioned the data into six smaller data sets based on the chr4 regions ( Figure 2 ) . This approach was necessary because heterochromatin replicates almost exclusively late , so any analysis that does not account for this fact merely compares heterochromatin to euchromatin . We then used a series of one-sample statistical tests to query whether probes with specific replication times were enriched or depleted for a specified genetic or epigenetic feature relative to the mean for that feature within a given region . This analysis is equivalent to comparing replication segments , but has the advantage of controlling for segment length by using probe numbers . Results for the proximal and distal portions of long arm are presented in Table 1 . ( The complete analysis is in Table S11 . ) In animal systems , early replication positively correlates with gene and GC content when integrated over large domains [7] , [12] , [13] , [62] . We found that the GC content of EM probes is depleted relative to the distal long arm , whereas the L probes are GC-enriched . EML probes have a GC content similar to EM probes , but M and ML probes are also GC-enriched . These results are likely linked to the gene coverage of these probes , with EM and EML probes showing depleted gene coverage and M , ML and L probes showing enriched gene coverage . The sequence content of the proximal long arm is different from the distal long arm , showing both a lower GC and gene content . However , the EM probes still show a lower GC content relative to the entire region . This depletion of gene and GC content in early-replicating regions contrasts with mammalian systems , and may reflect differences in genome structure . In both animals and plants , H3K4me is almost exclusively genic and correlates with gene expression [27] , [29] , [30] , [32] , [34] , [36] , [37] , with H3K4me3 having the strongest positive effect on gene expression in Arabidopsis [37] . H3K4me3 has been linked to early replication in mouse cells [13] , and all forms of H3K4me correlate with early replication in human cells [12] , [35] . We found that H3K4me1/2 is depleted in EM probes and enriched in ML and L probes in the distal long arm , consistent with the gene coverage . Despite its lower gene coverage , the proximal long arm has an abundance of H3K4me1/2 similar to that of the distal long arm , due in part to the gain of H3K4me1/2 by certain classes of TEs [34] . While we detected a depletion of H3K4me1/2 in EM probes , we did not detect a significant enrichment of H3K4me1/2 in L probes relative the proximal long arm as a whole . DNA cytosine methylation ( 5mC ) is found in the coding region of genes in the euchromatic regions of Arabidopsis , often in conjunction with H3K4me1 [33] , [34] , [37] , [38] , [63] . Like H3K4me1/2 , 5mC is depleted in EM probes and enriched in ML and L probes in the distal long arm . The distribution of 5mC differs between the proximal long arm and the distal long arm . While 88% of 5mC is genic in the distal long arm , the percentage drops to 60% in the proximal long arm , and much of the 5mC in this region is associated with TEs and other repetitive sequences located in heterochromatin [33] , [34] . We found a depletion of 5mC in EM , M and ML probes and an enrichment in L probes , which likely reflects the heterochromatic character of L probes in the proximal long arm . To confirm this hypothesis , we examined the distribution of histone H3K9me2 , which is associated with heterochromatin in Arabidopsis [31] , [34] , [64] . While H3K9me2 is not an abundant feature in the distal long arm , it is depleted in EM and M probes and enriched in L probes in this region , suggesting that some L probes are located in cryptic or facultative heterochromatin [28] , [31] , [65] . H3K9me2 is much more abundant in the proximal long arm , principally due to the elevated TE and repeat content of this region [33] , [34] , [58] . Again , H3K9me2 is depleted in EM , M and ML probes and enriched in L probes . The abundance of H3K9me2 , 5mC and late replication in the proximal long arm suggests that much of this region should be considered cryptic or facultative heterochromatin . Finally , we examined the correlation between H3K56ac and replication time . H3K56ac is associated with multiple biological processes that require an open chromatin conformation , including DNA replication , repair and transcription [40] , [66]–[71] . H3K56ac is enriched in gene promoter regions in Arabidopsis suggesting a role in transcription [34] . In both the distal and proximal long arms , we detected enrichment of H3K56ac in EM probes and depletion in L probes . H3K56ac is also enriched in EL probes in both the proximal and distal long arms and in EML probes in the proximal long arm . The enrichment of H3K56ac in regions depleted for genes and the epigenetic marks associated with genes raises the possibility that some of the H3K56ac detected in our cells may be related to DNA replication rather than gene transcription . To explore the relationship between genetic and epigenetic features and replication time in more detail , we performed further analyses in the context of the replicons identified above , again restricting our analysis to the long arm of chr4 . We compared the overall content of genetic features and epigenetic modifications between EM and L replicons . We found that gene coverage/content , GC content and H3K56ac are higher in EM than in L replicons , whereas L replicons are enriched for TEs , H3K9me2 and DNA 5mC ( Table 2 ) . H3K4me1/2 is similar in EM and L replicons ( Table 2 ) . While these results are more consistent with animal systems , the results for gene coverage , GC content and H3K4me1/2 seem to conflict with the probe-level analysis presented above . However , two factors must be considered . First , 62 of the 66 EM replicons are located in the distal long arm while 31 of the 42 L replicons are located in the proximal long arm , and the distal long arm has a higher gene content and GC content than the proximal long arm ( Table 1 ) . Second , many of the EM and L replicons are comprised of DNA segments that replicate in various parts of S phase , e . g . the termination zones of EM replicons often replicate in late S phase ( Figure 3 and Figure 4C ) . Thus , integration of genetic and epigenetic features over large regions such as replicons may obscure finer relationships . To further resolve these relationships , we devised an analysis that examined the distribution of features within an “average” replicon . A similar strategy was used to examine the distribution of epigenetic modifications across genes [33] , [34] . Each putative replicon in the proximal and distal long arms was divided into 10 intervals , each comprising 10% of its length . Unlike genes that have a definite polarity , most replicons are products of bidirectional fork progression and can be treated as symmetrical [1]–[5] . Hence , we combined our 10 intervals into 5 bins with the two innermost intervals near initiation zones comprising bin 1 and the two outermost intervals near termination zones comprising bin 5 . We determined the occurrence of gene-rich , AT-rich , H3K4me1/2 , H3K9me2 , H3K56ac and 5mC probes within each bin across EM and L replicons separately ( Figure 5 ) . We detected spatial correlations for both genetic and epigenetic features in EM replicons ( Figure 5 ) . Both AT-rich ( top 25% ) and H3K56ac probes are more abundant near initiation zones and depleted near termination zones ( Figure 5 ) . In contrast , the distribution of gene-rich ( top 25% ) , H3K4me1/2 and 5mC probes show opposite trends ( Figure 5 ) . H3K9me2 is sparse in EM replicons , and there is no spatial correlation ( Figure 5 ) . These results suggest that DNA replication initiates in AT-rich intergenic regions with an open chromatin conformation and proceeds by elongation into gene-rich regions where the epigenetic features associated with the gene regulation specify a more complex chromatin structure . Most of the spatial correlations do not apply to L replicons , although there is a clear enrichment of H3K56ac near initiation zones ( Figure 5 ) . This analysis reconciles the probe-level ( Table 1 ) and replicon analyses ( Table 2 ) , demonstrating that genetic and epigenetic features have both short and long range influences on replication time . To determine if the increased H3K56ac near initiation zones is linked with gene expression , we looked more closely at the relationship between replication time , gene expression and epigenetic modifications . Previous analysis of these cells showed that H3K56ac is enriched at the 5′ end and promoters of genes , while H3K4me1/2 and 5mC are enriched in the body of genes [34] . To discern broad patterns of epigenetic modification and gene expression , we generated heat maps of the epigenetic data using a loess algorithm as we did for replication time . We determined gene expression in our cells using existing microarray data [34] and used two metrics to measure gene expression . The presence/absence of a transcript was determined using the Affymetrix Micro Array Suite 5 . 0 algorithm ( MAS5 ) [72] . If the transcript was present , we considered the gene to be active . Gene expression levels were estimated using the gcRMA algorithm [73] . For the heat maps , we mapped the gcRMA expression values to the microarray probes prior to applying the loess algorithm . Representative late and early replicating regions of chr4 are shown in Figure 6 . Elevated levels of H3K56ac are frequently associated with regions near replicon initiation zones whereas elevated levels of H3K4me1/2 , H3K9me2 and 5mC are often near termination zones . Gene expression showed less clear-cut results sometimes colocalizing with H3K4me1/2 near termination zones and sometimes with H3K56ac near initiation zones . We then examined the effect of epigenetic modifications on gene expression and replication time at the level of genes . The 2844 chr4 genes with available expression and epigenetic data were classified into 16 groups based on the pattern of all possible combinations of the four epigenetic modifications examined in our cells [34] . Replication time for each gene was derived from the overlapping probes . Using MAS5 presence/absence calls , we estimated that 61% of chr4 genes are active in our cells . Using this as a baseline , we ranked the 16 epigenetic patterns by increasing gene activity , with genes displaying pattern 1 having the highest probability of expression and genes with pattern 16 having the lowest ( Table 3 ) . Genes with pattern 1 , which constitute the largest group , are positive for H3K4me1/2 , H3K56ac and 5mC ( Table 3 ) . The presence of H3K4me1/2 and 5mC on expressed genes is consistent with previous studies showing that these marks can potentiate gene expression in Arabidopsis [37] , [38] , [63] . Strikingly , H3K56ac is the only epigenetic modification found in all patterns that show increased gene activity ( Table 3 ) . A positive correlation between gene expression and H3K56ac has been shown in other organisms [66] , [70] , [71] , [74] , and we show that this correlation exists in Arabidopsis . For the remaining patterns , H3K9me2 showed a clear association with reduced gene activity while genes lacking detectable H3K4me1/2 , H3K9me2 , H3K56ac or 5mC also showed low activity ( Table 3 ) . Studies in other model organisms have shown a positive correlation between gene transcription and early replication [7]–[14] . When examined independent of epigenetic modifications , genes are significantly more likely to be expressed if they replicate EM rather than L ( Table 4 ) . Of chr4 genes , only genes with patterns 3 and 4 are more likely to replicate EM . Interestingly , genes with patterns 3 and 4 are distinguished from genes with patterns 1 and 2 by the lack of 5mC ( Table 3 ) . Despite their high frequency and levels of expression , genes with pattern 1 showed a slight tendency to replicate L and genes with pattern 2 showed no clear bias for either EM or L replication . Genes with patterns 7 , 14 and 15 are more likely to replicate L than EM , and each of these patterns is characterized by the presence of H3K9me2 and 5mC ( Table 3 ) . In summary , the increased expression of EM-replicating genes is associated with enrichment of this population for genes displaying H3K56ac but lacking 5mC as well as with depletion of genes bearing the repressive combination of H3K9me2 and 5mC . Allele-specific differences in replication time have been observed in animals [12] , [60] . This can occur when one allele of a gene bears activating epigenetic modifications while the other allele bears repressive modifications , and could give rise to EL , EML or I replication time . Genes with patterns 6 through 9 , 11 and 14 bear such modifications , and we did observe a slight enrichment of pattern 9 for EL genes and pattern 7 for EML genes ( Table 3 ) . However , the majority of the EL , EML and I segments cannot be explained by allele-specific replication timing . In many cases , genes that replicate EL , EML or I have only activating or repressive marks ( Table 3 ) . As stated above , many of these segments are associated with TEs and other repetitive elements . The heat maps suggested that much of the H3K56ac on chr4 is associated with early replication and not gene expression ( Figure 6 ) . To examine this more closely , we determined whether the H3K56ac near the initiation zones of replicons in the long arm of chr4 was due to genes with epigenetic patterns 1 through 4 or reflected H3K56ac in intergenic sequences as well . Genes with pattern 3 , positive only for H3K56ac , show a slight enrichment near initiation zones of EM replicons ( Figure S4 ) . An analysis of intergenic regions of chr4 revealed that the two most abundant epigenetic patterns are 3 ( H3K56ac only ) and 13 ( no detected modifications ) ( Table S12 ) . In the long arm of chr4 , intergenic regions with pattern 3 are enriched near initiation zones and depleted near termination zones , but intergenic regions with pattern 13 are uniformly distributed across replicons ( Figure S5 ) . To determine if this enrichment for intergenic H3K56ac near initiation zones is associated with the promoters of expressed genes , we analyzed the distribution of expressed genes ( regardless of epigenetic modifications ) across replicons . This analysis showed that expressed genes are uniformly distributed ( Figure S4 ) , allowing us to conclude that much of the intergenic H3K56ac is associated with early replication and not gene expression .
DNA replication has been profiled in Drosophila , mouse and human genomes [7]–[14] , [22] , [23] , [60] , [75] . Arabidopsis and Drosophila have a similar genome size ( ∼120 Mb each ) and gene density ( 250 and 111 genes/Mb respectively ) [54] , [57] , so it is not surprising that their replication profiles are similar . In contrast , the human and mouse genomes are substantially larger ( 3300 and 2500 Mb respectively ) and have a much lower gene density ( 10 genes/Mb each ) [55] , [56] . Mammalian genomes are also characterized by large regions of uniform GC and gene content known as isochores [62] , [76] , [77] . In both human and mouse cells , replication time has been shown to correlate with isochore structure , and high GC , gene-rich isochores tend to replicate early in S phase [13] , [62] . In contrast , it is not clear that a functionally equivalent isochore structure exists in Arabidopsis or Drosophila [77] , [78] . Such differences in genome structure may explain why gene content and expression and the associated epigenetic modifications have a more subtle influence on replication time in Arabidopsis than in mammals . For example , in human cells , distance to the closest expressed gene is strongly correlated with replication time [23] . However , these distances are on the order of megabases . Such a correlation is meaningless in Arabidopsis where the median intergenic distance is less than one kilobase [54] , [61] . Accordingly , we tailored our analysis to suit this compact genome , revealing many similarites and a few differences in the DNA replication programs of these model systems . A common approach to determine DNA replication timing utilizes the direct hybridization of BrdU labeled early and late S phase DNA to genomic tiling arrays to construct a replication profile that indicates the enrichment of a given sequence in early relative to late S phase [7]–[9] , [13] , [14] . In this approach , DNA replication in mid S phase is inferred rather than directly evaluated . We measured Arabidopsis DNA replication in early , mid , and late S phase cells in separate microarray experiments producing three independent replication profiles . This strategy revealed that the replication profiles for early and mid S phase cells are very similar to each other and clearly distinct from the late S phase profile . The majority of euchromatin in chr4 replicates in early and mid S phase , and the bulk of the heterochromatin replicates in late S phase ( Figure 2 ) . Temporal separation of DNA replication for euchromatin and heterochromatin was first observed at least five decades ago in both plants and animals [25] and is consistent with recent findings in Drosophila , mouse and human cells [21] . Fiber autoradiography experiments in Arabidopsis identified two temporal classes of replicons but did not distinguish euchromatin from heterochromatin [48] . Surprisingly , there is little difference between the early and mid S phase replication profiles ( Figure 2 ) , even though FACS profiles for the early and mid S phase cells are distinct ( Figure 1 and Figure S2 ) . When interpreting these results , it is important to remember that while FACS takes a DNA content measurement for each cell , the replication profiles are derived from a population of cells . If DNA replication followed a totally random program , a population of early S phase cells could produce a replication profile that encompasses the entire genome . At the other extreme is a strict temporal program in which the order of origin activation is highly consistent between cells in a population . With our experimental design , such a program would produce early S phase profiles showing replication of approximately 20% of the genome , while mid S phase profiles would be distinct from the early S phase profiles because they would encompass an additional 30% of the genome . Our results are an intermediate case between these two extremes and are best explained by a biphasic model of replication for Arabidopsis . In this model , the bulk of euchromatin replicates in early to mid S phase and the heterochromatin replicates late . Origin utilization is largely the same in early and mid S phase , suggesting that the temporal order of origin activation in the first half of S phase is stochastic . While we did not attempt to identify origins per se , we did identify initiation zones , and we detected few , if any , initiation zones specific to mid S phase cells ( Figure 3C and Figure 4B ) . The segmentation analysis showed some merging of early S phase segments to form larger mid S phase segments , but this effect most likely reflects elongation of replicons rather than activation of additional origins ( Figure 3 ) . The relative enrichment for initiation zones is similar in early and mid S phase cells suggesting that there is no quantitative difference in origin activation ( Figure 3C ) . In contrast , there are many putative initiation zones specific for late S phase ( Figure 3C and Figure 4B ) . The idea that DNA replication follows a strict temporal program derives largely from seminal work in budding yeast [6] , [79] . Budding yeast is characterized by sequence specific origins in a compact genome and , as such , might not be a good model for eukaryotes with much larger genomes and no clear origin sequence specificity [1]–[4] . Single molecule studies showed that even in budding and fission yeast , origin activation is stochastic and varies from cell to cell in a population [15] , [16] . Whole genome studies in Drosophila and mouse cells are also consistent with a biphasic model of DNA replication . In Drosophila , initiation zones are most abundant in early and late S phase [14] , while mouse replicons and replication domains tend to segment as either early or late [13] . Increasingly , origin activation is being interpreted as a largely stochastic process at the level of individual cells , with temporal profiles corresponding to the most probable sequence of origin activation for a population of cells [17] , [18] , [20] . The replication time of any given DNA segment is related to three factors – distance from the closest origin , activation time of that origin , and rate of DNA elongation upstream of the segment . Chromatin conformation can influence the latter two factors , and chromatin remodeling factors have been shown to be critical for DNA replication [80]–[83] . Our analyses of replication time with respect to both genetic and epigenetic features revealed correlations that may reflect the effect of chromatin conformation on origin specification , origin activity and the rate of DNA elongation . The heterochromatic knob and pericentromeric heterochromatin are entirely late replicating ( Figure 4A ) . Both of these regions are depleted in genes , rich in TEs , and display abundant H3K9me2 and 5mC ( Figure 2 and Table S11 ) . This constitutive heterochromatin exists in a compact conformation throughout most of the cell cycle [59] . This conformation likely restricts both origin activation and DNA elongation [2]–[4] . In both budding and fission yeast , pericentromeric heterochromatin replicates in early S phase [6] , [26] , but pericentromeric DNA replicates in late S phase in animal cells [9] , [84] , [85] . In both cases , replication of heterochromatin is dependent on chromatin remodeling complexes [80] , [82] , [83] , and it will be interesting to identify the complexes utilized by plants . We focused our analyses on the long arm of chr4 because it represents a large contiguous , genomic segment generally regarded as euchromatic [33] , [34] , [58] , [59] . However , we were surprised by the predominance of late replication in the proximal portion of the long arm ( Figure 2 and Figure 4A ) . Probe and replicon level analyses revealed that relative to the distal long arm , the proximal long arm has considerable heterochromatic character , including decreased gene coverage/content , increased TE coverage/content , and elevated levels of both H3K9me2 and DNA 5mC ( Figure 2 , Table 1 , and Table 2 ) . Much of the proximal long arm likely adopts a chromatin state known as cryptic or facultative heterochromatin [28] , [65] . Such regions share some of the biochemical features of constitutive heterochromatin , including hypoacetylation , H3K9me2 and DNA 5mC , but do not adopt the long range , highly condensed structure of constitutive heterochromatin . In mouse cells , replication domains that switch replication time upon differentiation are believed to be facultative heterochromatin [13] , [21] . Despite the overall differences in replication time for the proximal long and distal long arm regions , we detected several correlations between replication time and genetic and epigenetic features that were similar in both regions . For example , EM-replicating probes show increased AT content , decreased gene coverage and decreased DNA 5mC ( Table 1 ) . Further , the histone modifications , H3K4me1/2 and H3K9me2 , are decreased while H3K56ac is increased . The pattern is opposite for L-replicating probes . These observations suggest that DNA replication initiates in AT-rich intergenic regions with an open chromatin conformation and proceeds into regions where the epigenetic modifications associated with gene expression specify a more complex chromatin conformation . The distribution of genetic and epigenetic features within replicons further supports this hypothesis ( Figure 5 ) . The EM replicons display trends that are consistent with a replicon model that has been termed the “relaxed replicon” model [2]–[4] . This model incorporates several mechanisms to explain ORC binding and replicon structure in higher eukaryotes . Mechanisms consistent with our work include a higher affinity of ORC for open chromatin and AT rich sequences [86] , [87] , transcriptional interference preventing ORC binding [88] , and inhibition of ORC binding by DNA methylation [89] . The structure of EM replicons may be driven by the probability of both ORC binding and origin activation . Regions proximal to initiation zones have a higher AT content and elevated H3K56ac and may have a higher probability of binding ORC to form an origin ( Figure 5 ) . The lower gene density , lower H3K4me1/2 and reduced 5mC in these regions would also favor origin formation . Termination zones show opposite trends for these characteristics , consistent with a lower probability of binding ORC . In addition , elevated levels of H3K4me1/2 and 5mC may impede the progress of replication forks in these regions . Chromatin modified by DNA 5mC adopts a more compact conformation and impedes the progress of RNA polymerase [63] , [90] . The trends for EM replicons are readily apparent when the epigenetic modifications are integrated over large regions ( Figure 6 ) . Most of the trends do not hold for L replicons , which in comparison to EM replicons , have greatly elevated and evenly distributed levels of H3K9me2 and 5mC indicative of a heterochromatic state ( Table 2 , Figure 5 , and Figure 6 ) . Replication may be delayed in these regions because it requires the activity of chromatin remodeling complexes , as discussed above for the heterochromatic knob and pericentromere . Additionally , L replicons may have a lower density of potential origins . Gene expression shows a positive correlation with early replication in all higher eukaryotes examined to date [7]–[9] , [11]–[14] , [22] , [23] . This correlation is strongest when integrated over large regions because there are many exceptions at the level of individual genes . We identified a similar correlation in Arabidopsis , with genes in EM replicating regions more likely to be expressed than genes in L replicating regions ( Table 4 ) . However , the relationship of specific epigenetic modifications to gene expression and replication time is complex ( Table 3 ) . From the standpoint of replication time , two effects are prominent . H3K56ac with a lack of H3K9me2 is favorable for both gene expression and early replication , whereas H3K9me2 with a lack of H3K56ac correlates with lower expression and late replication . Genes associated with both H3K9me2 and H3K56ac also tend toward low expression and late replication , but the effect is less clear-cut than H3K9me2 alone . Genes with both marks are similar to the “pan S” or “biphasic” genes in human cells which bear both active and repressive chromatin marks due to interallelic variation [12] , [60] . We also observed an increase in EML replication for these genes in Arabidopsis ( Table 3 ) . Unlike the epigenetic modifications discussed above , integration of gene expression over large regions did not reveal a correlation between gene expression and replicon structure ( Figure 6 ) . This lack of correlation probably reflects the fact that the expression of an individual gene is more strongly modulated by epigenetic modifications specific to that gene rather than by the global characteristics of large regions containing many genes . H3K56ac is thought to occur on all newly synthesized H3 histones and be required for nucleosome assembly [66] , [69] , [91] . H3K56ac is associated with regions of nucleosome exchange such as active promoters [70] , [71] , sites of DNA repair [67] , [69] , and nascent chromatin [40] , [68] . In budding yeast , H3K56ac is most abundant during S phase and localizes to early origins in a cell cycle dependent manner [40] , [68] , [69] . Intriguingly , H3K56ac correlated with EM replication and was enriched at the center of Arabidopsis replicons ( Figure 5 and Figure 6 ) . Interpretation of this data must be tempered by the fact that the epigenetic profiling was performed on an unsorted population of cells so both replication dependent and independent H3K56ac is represented . Although there was a positive correlation between H3K56ac and gene expression ( Table 3 ) , integration of H3K56ac over large regions , including intergenic regions , showed a clear association with replication time and not with gene expression ( Figure 6 ) . H4K16ac correlates with early replication in Drosophila [14] , while H3K56ac associates with early origins in budding yeast [40] . We have provided the first evidence , to our knowledge , linking H3K56ac to replication time in a higher eukaryote . Unlike H4K16 , H3K56 is located in the core of the histone and is inaccessible to acetylation in the context of a fully assembled nucleosome [66] , [92] . Therefore , H3K56ac might be associated with nascent DNA behind active replication forks rather than the disassembly of chromatin ahead of replication forks [68] . Nevertheless , H3K56ac may prove to be a valuable epigenetic mark for identifying replication origins . We have presented a high-resolution analysis of the replication program for a plant chromosome . Arabidopsis DNA replication is biphasic , with euchromatin replicating in the first half of S phase and heterochromatin replicating in the last half . This pattern is similar to other eukaryotes [9] , [84] , [85] , although exceptions do occur in yeast [6] , [26] . Within each phase , origin activation appears to be largely stochastic because we could discern few differences between replication profiles for early and mid S phase cells . This result provides additional support for the emerging model of stochastic origin activation rather than strict temporal regulation [15]–[18] , [20] . The replication profiles allowed us to construct a replicon map for chr4 and to correlate replication time with gene expression and specific epigenetic modifications . We showed that initiation zones are enriched for epigenetic features associated with open chromatin , providing support for the “relaxed replicon” model , which proposes that origin specification and activity are strongly influenced by both sequence content and chromatin conformation in higher eukaryotes [1]–[4] . Finally , we showed that early replicating regions and initiation zones are enriched for H3K56ac . This histone modification continues to be an area of intense research because of its role in DNA replication , DNA repair and gene expression . We provide evidence that H3K56ac has both replication independent and dependent roles in plants by showing that genes bearing H3K56ac have a higher probability of expression , whereas large regions with elevated H3K56ac levels are associated with early replication . Replication time and H3K56ac data in conjunction with other experiments may help us identify replication origins in plants . This study linking DNA replication and replicon structure to chromatin conformation provides a foundation for future studies that will investigate the impact of these processes on plant growth and development .
The Arabidopsis cell line ( Col-0 , ecotype Columbia ) was maintained in Gamborg's B5 basal medium with minor salt ( Sigma G5893 ) supplemented with 1 . 1 mg/L 2 , 4-dichlorophenoxyacetic acid , 3 mM MES and 3% sucrose . The cells were grown on a rotary shaker at 160 rpm under constant light at 23°C and subcultured every 7 days with a 1∶10 ( inoculum∶fresh medium ) dilution [34] . BrdU labeling was maximized using a ‘7-d split culture' by mixing 25 mL of fresh medium and 25 mL of the Arabidopsis culture at 7 days post subculture . The 7-d split culture was grown for 16 h and then labeled for 1 h with 100 µM BrdU ( Sigma B9285 ) . Labeled cells were fixed in 1% paraformaldehyde for 15 min , washed in 1× phosphate buffered saline ( PBS ) three times , and snap frozen in liquid nitrogen . Time course experiments showed that BrdU incorporation is highest between 12 and 16 h post-labeling ( Figure S1C ) . Cells from six 7-d split cultures were combined for each biological replicate . The frozen cell pellet for each biological replicate was ground at 4°C in 150 mL lysis buffer ( 15 mM Tris-HCl pH 7 . 5 , 2 mM EDTA , 80 mM KCl , 20 mM NaCl , 15 mM β-mercaptoethanol , and 0 . 1% Triton X-100 ) using a commercial blender . The ground cell suspension was incubated at 4°C for 5 min and filtered through a 3-tiered nylon mesh ( 100 , 50 , and 30 µm ) . The filtrate was centrifuged at 200 ×g for 5 min at 4°C , and the nuclei were resuspended in 8 mL of lysis buffer containing 2 µg/mL DAPI and 50 µg/mL RNase A . The isolated nuclei were filtered through a 20-µm nylon filter before flow cytometric analysis and sorting . Nuclei were sorted and recovered using an InFlux cell sorter ( BD Biosciences ) equipped with a 355-nm UV laser and a 488-nm sapphire laser . STE buffer ( 10 mM Tris-HCl pH 7 . 5 , 1 mM EDTA , and 100 mM NaCl ) was used as a sheath fluid , and nuclei were sorted into a 50-mL tube containing 5 mL STE buffer . An analytical FACS profile for BrdU incorporation and DNA content was generated as described [93] with some modifications . BrdU-labeled cells were fixed in 70% ethanol on ice for 1 h and frozen in liquid nitrogen . Nuclei were isolated , denatured in 2N HCl and 0 . 5% Triton X-100 at room temperature for 30 min , neutralized by adding 0 . 1 M Na2B4O7 ( pH 8 . 5 ) , and washed twice with PBS-TBR ( 1x PBS , 1% BSA , 0 . 5% Tween-20 and 50 µg/mL RNase A ) . The nuclei were resuspended in PBS-TBR containing a 1∶50 dilution of anti-BrdU Alexa Fluor 488 conjugate ( Invitrogen ) by gentle agitation overnight at 4°C in the dark . The nuclei were washed once with PBS-TBR , incubated in PBS-TBR containing 10 µg/mL propidium iodide for at least 30 min , filtered through a 20-µm nylon filter , and analyzed by FACS . FlowJo ( Version 8 . 8 . 6 ) software was used for the data analysis . To reverse the crosslinks , the sorted nuclei were treated with 50 mM EDTA , 1% sodium lauroyl sarcosine and 200 µg/mL proteinase K for 1 h at 42°C and then overnight at 65°C in the dark . The mixture was supplemented with 4 mg/mL phenylmethanesulphonylfluoride and incubated for 40 min at room temperature prior to extraction of genomic DNA using phenol/chloroform/IAA in a phase lock gel ( Sigma ) . The upper aqueous phase was mixed with 150 µg/mL GlycoBlue ( Ambion ) and precipitated with 0 . 3 M sodium acetate and 2 volumes of cold ethanol . The DNA was centrifuged and the pellet was washed with 70% ethanol once , dried for 5 min using a SpeedVac in the dark , and resuspended in sterile water . BrdU-labeled DNA was immunoprecitated as described [94] with some minor modifications . Genomic DNA extracted from the sorted nuclei was sonicated in 450 µL of ChIP dilution buffer ( 0 . 1% BSA , 1 . 2 mM EDTA , 16 . 7 mM Tris-HCl pH 8 , and 167 mM NaCl ) to a shear-size of 500 to 1000 bp , followed by addition of Triton X-100 ( 1 . 1% ) . The sheared DNA was denatured at 95°C for 5 min and immediately cooled on ice for at least 5 min . One mL of cold ChIP dilution buffer containing 1 . 1% Triton X-100 was added and the sheared DNA was incubated with 0 . 5 µL anti-BrdU antibody ( Invitrogen ) for 3 h at 4°C . DNA containing BrdU was immunoprecipitated by adding 100 µL of 50% protein G-sepharose beads ( Sigma ) and incubating overnight in the dark at 4°C with gentle agitation . The beads were washed as previously described by Gendrel , et al . ( 2005 ) . BrdU-labeled DNA was eluted from the beads with 0 . 2 M glycine ( pH 2 . 5 ) and neutralized by adding 10% ( v/v ) of 1 M Tris-HCl ( pH 8 ) . Eluted DNA was treated with proteinase K for 1 h at 45°C , extracted with phenol/chloroform/IAA , and precipitated with sodium acetate and ethanol . Precipitated DNA was resuspended in RT-PCR grade water ( Ambion ) and used as template for random amplification and real-time quantitative PCR . BrdU immunoprecipitated DNA ( target DNA ) and input DNA ( reference DNA ) samples were amplified as described [95] , purified and concentrated to 200–250 ng/µL using a QIAquick PCR Purification Kit ( QIAGEN ) . Each amplified DNA sample ( 1 . 5 mg ) was labeled with either Cy3 or Cy5 fluorescent dye and purified using a BioPrime Array CGH Genomic Labeling System ( Invitrogen ) . The Cy dye-labeled target and reference samples were co-hybridized on a custom-printed tiling array [33] with a dye-swap experimental design . Each experiment comprised six microarrays representing the three biological replicates and the corresponding dye swaps . Microarray hybridization and washing were previously described [34] but modified to include DyeSaver2 coating reagent ( Genisphere ) to minimize oxidation of Cy5 . Hybridized microarrays were scanned using a PerkinElmer ScanArray Express scanner and quantified using GenePix Pro software ( version 6 . 01 ) . Calculation of microarray probe enrichment ratios , loess and quantile normalizations were done in the R statistical computing environment with the limma package using default settings [96]–[98] . Probe ratios were loess-smoothed in a 150-kb window for replication profiles and identification of initiation and termination zones . Segments of contiguous replication time were defined as regions where smoothed probe ratios were greater than zero for a minimum of 10-kb . This filter minimized excessive replication time changes in regions with low probe enrichment ratios . Merging of the segmentations for early , mid and late S phase cells was done by determining the regions of overlap . The 10-kb length minimum was not used at this step . Initiation and termination zones were identified as the inflection points of the loess-smoothed profiles as described in the results . Zones were then defined as the 10-kb regions centered at the inflection point . Overlapping zones were merged into a single zone . Replication boundaries were chosen from the three sets of termination zones based on the following order of precedence: 1 ) termination zones present in early , mid and late S phase cells , 2 ) termination zones enriched in late S phase and 3 ) termination zones that manifest as local minima but enriched in early and/or mid S phase . All data manipulation and statistical analysis was performed with R and Bioconductor [96] , [99] . A database incorporating probe ratios for replication time , histone modifications , DNA 5mC and the TAIR8 Arabidopsis genome annotation [61] was constructed to facilitate analysis . Gene and TE coverage values for probes and larger regions are the percentage of bases in that region that overlap with any gene or TE respectively . Overlapping genes or TEs were treated as one so that coverage values do not exceed 100% . Statistical comparisons of GC content and gene or TE coverage were performed by one-sample t-tests . AT-rich and gene-rich probes were defined as the top quartile of all probes on the array . AT-rich , gene-rich and probes positive for histone modifications or DNA methylation data were treated as binomial data , and a one-sample binomial test was used for analyses . Gene expression values were determined using the affy package in R [73] . MAS5 presence or absence calls and gcRMA expression values were calculated using default settings . The pattern of epigenetic modifications for chr4 genes was determined from the modifications of the overlapping probes again treating the modifications as binomial data . Heat maps for epigenetic modifications were generated by smoothing probe ratios in a 150-kb window as for replication profiles and ranking the data by deciles for the whole of chr4 . Heat maps for gene expression were generated similarly but gcRMA expression values were used rather than probe ratios . R scripts for all analyses and figures are available upon request . | During growth and development , all plants and animals must replicate their DNA . This process is regulated to ensure that all sequences are completely and accurately replicated and is limited to S phase of the cell cycle . In the cell , DNA is packaged with histone proteins into chromatin , and both DNA and histones are subject to epigenetic modifications that affect chromatin state . Euchromatin and heterochromatin are chromatin states marked by epigenetic modifications specifying open and closed conformations , respectively . Using the model plant Arabidopsis thaliana , we show that the time at which a DNA sequence replicates is influenced by the epigenetic modifications to the surrounding chromatin . DNA replication occurs in two phases , with euchromatin replicating in early and mid S phase and heterochromatin replicating late . DNA replication time has been linked to gene expression in other organisms , and this is also true in Arabidopsis because more genes are active in euchromatin when compared to heterochromatin . The earliest replicating DNA sequences are associated with acetylation of histone H3 on lysine 56 ( H3K56ac ) . H3K56ac is also abundant in active genes , but the patterns of association of H3K56ac with gene expression and DNA replication are distinct , suggesting that H3K56ac is independently linked to both processes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/genomics",
"molecular",
"biology/dna",
"replication",
"molecular",
"biology/chromatin",
"structure",
"molecular",
"biology/histone",
"modification",
"biochemistry/replication",
"and",
"repair",
"molecular",
"biology/dna",
"methylation",
"genetics",
... | 2010 | Arabidopsis thaliana Chromosome 4 Replicates in Two Phases That Correlate with Chromatin State |
Tapeworm ( cestode ) infections occur worldwide even in developed countries and globalization has further complicated the epidemiology of such infections . Nonetheless , recent epidemiological data on cestode infections are limited . Our objectives were to elucidate the clinical characteristics and epidemiology of diphyllobothriosis and taeniosis in Tokyo , Japan . We retrospectively reviewed 24 cases of human intestinal cestode infection from January 2006 to December 2015 at a tertiary referral hospital in Tokyo , Japan . The patients included were diagnosed with cestode infection based on morphological and/or molecular identification of expelled proglottids and/or eggs and treated in our hospital . Fifteen and 9 patients were diagnosed with diphyllobothriosis and taeniosis , respectively . The median patient age was 31 years ( interquartile range [IQR]: 26–42 years ) , and 13 ( 54% ) were male . Most of the patients ( 91 . 7% ) were Japanese . All patients were successfully treated with praziquantel without recurrence . Diphyllobothriosis was caused by Diphyllobothrium nihonkaiense in all patients . Taeniosis was due to infection of Taenia saginata in 8 [88 . 9%] patients and T . asiatica in 1 [11 . 1%] patient . All patients with taeniosis were infected outside Japan , as opposed to those with diphyllobothriosis , which were domestic . The source locations of taeniosis were mostly in developing regions . The median duration of the stay of the patients with taeniosis at the respective source location was 1 month ( IQR: 1–8 ) . The cestode infection , especially with D . nihonkaiense , has frequently occurred , even in Japanese cities , thereby implicating the probable increase in the prevalence of diphyllobothriosis among travelers , as the number of travelers is expected to increase owing to the Tokyo Olympics/Paralympics in 2020 . In addition , medical practitioners should be aware of the importance of providing advice to travelers to endemic countries of taeniosis , including the potential risks of infection and preventive methods for these infections .
Tapeworm ( cestode ) infections , including diphyllobothriosis and taeniosis , occur worldwide even in developed countries . Causative species of cestode infections differ significantly depending on the geographical areas . Diphyllobothriosis affects an estimated 10–20 million worldwide [1 , 2] . Among diphyllobothriosis , Diphyllobothrium latum infection due to ingesting undercooked freshwater fish ( such as pike , perch , or rainbow trout ) is common in Europe , Russia , and South America [3 , 4] . However , in Japan , Diphyllobothrium nihonkaiense infection associated with consuming raw Pacific salmon is the most prevalent fishborne cestodiasis . Diphyllobothriosis is usually asymptomatic , but anemia due to vitamin B12 deficiency ( especially in D . latum infection ) , abdominal symptoms ( i . e . , pain , discomfort , and diarrhea ) , weight loss , and dizziness have been reported [1–4] . Taeniosis is caused by Taenia saginata , Taenia solium , and Taenia asiatica infections in humans and is associated with ingesting raw or undercooked beef or pork . T . saginata and T . solium infections are found worldwide , while T . asiatica infections are limited to Asia . Even in Japan , autochthonous T . asiatica infections have been sporadically reported after 2010 [5] . Similar to diphyllobothriosis , although symptoms , such as abdominal cramps and malaise , are described occasionally , symptoms are usually absent . Globalization has further complicated the epidemiology of cestode infections [6] . Transportation of causative food items , as well as traveling to endemic areas , has caused a rise in cestode infections outside previously known endemic areas [7–9] . For example , in Japan , endemic areas of diphyllobothriosis were previously limited to the coastal region on the Sea of Japan; however , diphyllobothriosis cases have been reported across Japan these days . Nonetheless , recent epidemiological data on cestode infections are limited . The present study aimed to elucidate the clinical characteristics and epidemiology of diphyllobothriosis and taeniosis in Tokyo , Japan .
This study was approved by the Human Research Ethics Committee of NCGM ( NCGM-G-001994-00 ) and all data analyzed were anonymized .
Fifteen and 9 patients were diagnosed with diphyllobothriosis and taeniosis , respectively . The median patient age was 31 years ( interquartile range [IQR]: 26–42 years ) , and 13 ( 54% ) were male ( Table 1 ) . Most of the patients ( 91 . 7% ) were Japanese . All patients were treated with praziquantel ( 600 mg once daily , except 1 with D . nihonkaiense treated with 1500 mg based on decision by an attending physician ) . No patient experienced recurrence . Diphyllobothriosis was caused by D . nihonkaiense in all patients . Fourteen ( 93 . 3% ) patients presented with histories of consuming raw salmon . Prepatent periods could not be estimated in patients with diphyllobothriosis because most of them regularly consumed raw fish . In 9 of 15 ( 60% ) cases , the diagnosis was confirmed through cestode cox1 gene sequencing . Four ( 26 . 7% ) patients presented with abdominal symptoms , such as pain/discomfort and diarrhea; one ( 6 . 7% ) exhibited weight loss . No patient had eosinophilia . After treatment , scoleces were detected in only 6 ( 40% ) patients . Taeniosis was due to infection of T . saginata in 8 ( 88 . 9% ) patients and T . asiatica in 1 ( 11 . 1% ) patient; these patients had consumed raw beef and pork liver , respectively . The median prepatent period of taeniosis was 4 months ( IQR: 1 . 5–5 ) . In 7 of 9 ( 77 . 8% ) cases , the diagnosis was confirmed via cestode cox1 gene sequencing . Three ( 33 . 3% ) patients presented with abdominal symptoms; no patient exhibited weight loss . Eosinophilia was detected in only 1 patient with T . saginata infection of 5 ( 20% ) patients with taeniosis for whom laboratory test results were available . After treatment , scoleces were detected in only 2 ( 22 . 2% ) patients . Patients with taeniosis were slightly older than those with diphyllobothriosis , and tended to display a longer duration between symptom onset and the first clinical visit ( Table 2 ) . All patients with taeniosis were infected outside Japan , as opposed to those with diphyllobothriosis , which were domestic . The source locations of taeniosis infection were mostly in developing regions such as Africa , Southeast Asia , or the Middle East . The median duration of the stay of the patients with taeniosis at the respective source location was 1 month ( IQR: 1–8 ) .
This study highlights that the cestode infection , especially with D . nihonkaiense , has frequently occurred , even in Japanese cities . In Japan , 114 cases infected with D . nihonkaiense from all over the country were identified at the National Institute of Infectious Diseases ( NIID ) , Tokyo , Japan , and 325 cases were reported on the medical articles between January 2007 and March 2017 [12] . These facts mean that approximately 40 cases of autochthonous D . nihonkaiense infection were annually reported over the last decade in Japan . However , the actual number is expected to be much higher and it would probaply be several times as many as reported number annually because of the cases diagnosed at the other institute or unreported [12] . As traditional Japanese meals and restaurants have become increasingly popular worldwide and there are no regulation that recommends to freeze fishes before consuming it raw in Japan , like that of the US Food and Drug Administration [13] , it would be valuable to advise people not only Japanese consumers and manufacturers but also to travelers from foreign countries about the potential risks of eating raw salmon without freezing . In the present study , diphyllobothriosis was caused by D . nihonkaiense in all patients , who were presumed to be infected in Japan . These results were consistent with the fact that D . latum infection , confirmed through molecular analysis , from humans has not been reported in Japan [14] . Determining the geographical area of infection is helpful for identifying the cestode species . The estimated infection sites were significantly different between taeniosis and diphyllobothriosis . In the present study , the finding that all taeniosis infections occurred in foreign , mainly developing , countries reflects the fact that , in Japan , taeniosis has mainly been reported among travelers to or from endemic areas . From January 1990 to March 2017 , eighty-eight cases of taeniosis from all over the country were identified at the NIID and 95 cases were published on medical articles . Based on these data , approximately seven cases with taeniosis have been reported annually in recent years in Japan [12] . Along with diphyllobothriosis , the actual number of taeniosis is expected to be much higher because the cases of taeniosis should have been under-reported [12] . In addition , although imported T . solium , including cysticercosis , and domestic T . asiatica infections were not identified in this study , they can occur in Japan [5] . For instance , in 2010 , nineteen out of twenty patients infected with T . asiatica were Japanese nationals residing in the Kanto region where Tokyo belongs to , and fifteen patients stated that they frequently ate raw pig liver . In addition , sixteen of them had never been overseas or , if they had undertaken any international travel , they traveled to countries where T . asiatica is not endemic [5] . This study showed two other important findings . First , the lengths of the stay in endemic areas in taeniosis cases were relatively short . This finding highlights the importance of pre-travel education on appropriate food selections for those planning to stay in endemic countries , even for short periods . Second , patients with taeniosis were significantly older to those with diphyllobothriosis , probably because middle-aged people , rather than younger people , are likely to travel abroad for much longer periods for work purposes . In the present study , 6 out of 9 patients with taeniosis visited endemic areas for work . D . nihonkaiense infection often occurs even in cities in Japan , thereby implicating the probable increase in the prevalence of diphyllobothriosis among travelers , as the number of travelers is expected to increase owing to the Tokyo Olympics/Paralympics in 2020 . In addition , considering the increasing number of travelers to foreign countries where taeniosis is endemic , medical practitioners should be aware of the importance of providing advice to travelers including the potential risks of infection , and preventive methods for these infections should be considered . | Tapeworm ( cestode ) infections occur worldwide even in developed countries . Causative species of cestode infections differ significantly depending on the geographical areas and globalization has further complicated the epidemiology of such infections . Nonetheless , recent epidemiological data on cestode infections are limited . Our objectives were to elucidate the clinical characteristics and epidemiology of diphyllobothriosis and taeniosis in Tokyo , Japan . This study revealed that the cestode infection , especially with D . nihonkaiense , has frequently occurred , even in Japanese cities , thereby implicating the probable increase in the prevalence of diphyllobothriosis among travelers , as the number of travelers is expected to increase owing to the Tokyo Olympics/Paralympics in 2020 . In addition , compared to diphyllobothriosis , taeniosis was occurred mainly among travelers to and from developing countries even for short period of stay ( median 1 month [IQR:1–8] ) . Medical practitioners should be aware of the importance of providing advice to travelers to endemic countries of taeniosis , including the potential risks of infection and preventive methods for these infections . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"beef",
"medicine",
"and",
"health",
"sciences",
"fish",
"pathology",
"and",
"laboratory",
"medicine",
"cestodes",
"helminths",
"japan",
"geographical",
"locations",
"diet",
"vertebrates",
"parasitic",
"diseases",
"animals",
"animal",
"products",
"ostei... | 2018 | Clinical characteristics and epidemiology of intestinal tapeworm infections over the last decade in Tokyo, Japan: A retrospective review |
The Republic of Haiti is one of only several countries in the Western Hemisphere in which canine rabies is still endemic . Estimation methods have predicted that 130 human deaths occur per year , yet existing surveillance mechanisms have detected few of these rabies cases . Likewise , canine rabies surveillance capacity has had only limited capacity , detecting only two rabid dogs per year , on average . In 2013 , Haiti initiated a community-based animal rabies surveillance program comprised of two components: active community bite investigation and passive animal rabies investigation . From January 2013 –December 2014 , 778 rabies suspect animals were reported for investigation . Rabies was laboratory-confirmed in 70 animals ( 9% ) and an additional 36 cases were identified based on clinical diagnosis ( 5% ) , representing an 18-fold increase in reporting of rabid animals compared to the three years before the program was implemented . Dogs were the most frequent rabid animal ( 90% ) . Testing and observation ruled out rabies in 61% of animals investigated . A total of 639 bite victims were reported to the program and an additional 364 bite victims who had not sought medical care were identified during the course of investigations . Only 31% of people with likely rabies exposures had initiated rabies post-exposure prophylaxis prior to the investigation . Rabies is a neglected disease in-part due to a lack of surveillance and understanding about the burden . The surveillance methods employed by this program established a much higher burden of canine rabies in Haiti than previously recognized . The active , community-based bite investigations identified numerous additional rabies exposures and bite victims were referred for appropriate medical care , averting potential human rabies deaths . The use of community-based rabies surveillance programs such as HARSP should be considered in canine rabies endemic countries .
Rabies is the deadliest of all zoonotic diseases , responsible for more than 59 , 000 human deaths , annually [1 , 2] . It is also the most lethal infectious disease , with a case fatality rate of nearly 100% even with advanced medical intervention [1] . Although all mammals are susceptible to rabies virus infection , only certain reservoir species are capable of maintaining enzootic circulation through conspecific transmission . Reservoir species for the rabies virus include bats in the Western Hemisphere and at least 15 terrestrial mammals worldwide . While all reservoir species can transmit rabies to people , none are more significant than the domesticated dog , which is responsible for nearly all human rabies deaths [2] . In most of the developing world , rabies surveillance methods are ineffective , resulting in underreporting of human and animal cases [3] . The costs of developing and implementing a comprehensive surveillance system for rabies are prohibitive in many developing countries . The methods of rabies surveillance practiced in many countries suffer from fundamental problems including a lack of trained professionals and a lack of diagnostic laboratory capacity capable of providing results clinically relevant for human exposures [3 , 4] . In the absence of systems capable of timely investigation and reporting for clinical and public health purposes , there are often few perceived incentives for conducting rabies surveillance . This results in a lack of awareness of case burden , reduced funding for control , and poor community engagement around prevention [3 , 4] . This is often referred to as the ‘cycle of neglect’ and contributes to the poor understanding of the global canine rabies burden as well as the continued lack of funding for its control and prevention . Due to this dearth of surveillance data , estimation methods must be periodically conducted to understand the national and global scope of rabies burden [2] . Multiple burden estimation methods have been attempted , all of which have placed Haiti amongst the highest for canine and human rabies in the Western Hemisphere [2 , 5–7] . However , inferences from these global methods on disease burden at fine geographical resolution are limited by poor data availability . Routine passive rabies surveillance systems can provide this region-specific data which integral to inform rabies control activities . Surveillance for rabies provides additional benefits , including the removal of rabid animals from the community by trained professionals , reduction of rabies exposures , and evidence-based use of rabies biologics [8] . Canine rabies has been eliminated in most developed countries , largely due to successful rabies vaccination programs and responsible dog population management [1 , 4 , 9 , 10] . The most recent reduction in the burden of canine rabies has been observed in the Western Hemisphere , where it reached historically low levels as of 2014 [6] . The Republic of Haiti , a Caribbean nation of 10 . 5 million people , has been identified as one of only several countries in the Western Hemisphere in which these advances in canine rabies control have not been mirrored . The true incidence of human and canine rabies in Haiti is currently not known , however , limited surveillance activity from 2009–2012 detected an average of four canine and seven human rabies cases annually [6 , 7] . In 2011 , a laboratory-based animal rabies surveillance program was developed to improve our knowledge of the canine rabies burden in Haiti and enact community-level preventive measures . This report details the process of developing the program , data of animal rabies cases and bites reported through this program , and a comparison to the pre-surveillance period of 2009–2012 .
Prior to 2012 , all rabid animals in Haiti were diagnosed by Seller’s stain , a method with lower sensitivity and specificity compared to the international standard Direct Fluorescent Antibody test ( DFA ) . During a 24 month period from 2011 through 2012 CDC assisted in the establishment of an animal rabies diagnostic facility at the MARNDR—National Veterinary Laboratory . The laboratory was established in a dedicated room within the virology unit and was outfitted with one fluorescent microscope , an incubator , freezer , fume hood , and supplies required for processing and diagnosing samples . Staff were trained in the method of DFA and Direct Rapid Immunohistochemistry Test ( DRIT ) tests . Proficiency of testing was conducted through confirmatory testing of samples at CDC and through participation in the Latin American diagnostic proficiency testing program . In December of 2012 and June of 2013 trainings were held on principles of rabies education and prevention . Trainings encompassed both classroom and field coursework . Eighteen participants were selected by MARNDR for training and had a background in veterinary education at either the para-professional ( nine-week training program ) or technician level ( two-year degree ) . In January 2013 , one veterinary technician began surveillance activities under HARSP in the Petionville commune , within the West Department . In June 2013 , three veterinary para-professionals joined HARSP . The four selected animal rabies surveillance officers ( ARSO ) received rabies pre-exposure vaccination , training in animal rabies surveillance and bite investigation , and were provided equipment for safe and humane animal capture . Rabies is only detectable by current laboratory methods after the virus has infected neurons in the brain stem and cerebellum and it is during this time that the animal will show signs of illness . Therefore , passive surveillance efforts targeting animals which are involved in a bite event or have clinical signs are more likely to detect rabid animals . In Haiti , bites recorded at sentinel hospitals are notifiable events to MSPP–Department of Epidemiology and Laboratory Research; therefore HARSP relies upon inter-ministerial collaborations for animal surveillance in which MSPP reports bites daily to MARNDR for investigation . Summary reports of rabies investigations are shared at weekly One Health meetings . The HARSP also utilizes community , non-sentinel hospital , and veterinary-based reporting of rabies suspect animals for investigation ( Fig 1 ) . No formal media campaigns were enacted to advertise HARSP to these stakeholders , rather they were informed through word-of-mouth and engagement during community-based investigations . Haiti is administratively comprised of ten departments which are divided into 140 communes [11] . Four ARSOs were employed through MARNDR to operate primarily within three communes in the West Department , the most developed of the ten departments; Petionville , Carrefour , and Croix-des-Bouquets . Communes were selected based on proximity to the MARNDR National Veterinary Laboratory and adequate road infrastructure to facilitate travel during investigations . Petionville and Carrefour are communes adjacent to Port-au-Prince . Petionville is considered an affluent commune with a population of 342 , 694 people [12] . Carrefour is an impoverished commune with a population of 465 , 019 people [12] . Croix-des-Bouquets is located 13 miles north from the capital city and has a population of 84 , 812 people [12] . The overall population size targeted by HARSP was c . 1 million ( or 8 . 5% of Haiti’s population ) . These three communes were the centers for operations for ARSO’s , but investigations were conducted in other communes as time and resources allowed . HARSP investigations occurred during a two-year surveillance period ( January 2013 to December 2014 ) . Investigations were composed of two components: community bite investigation and animal rabies investigation . Animals that appeared healthy were placed on a 14-day in-home observation with both verbal and written instruction on proper animal care and rabies exposure prevention ( animal quarantine facilities are not available in Haiti ) . Animals healthy after 14 days were released from observation . Animals that displayed signs consistent with rabies during the investigation or the observation period were anesthetized and euthanized according to American Veterinary Medical Association standards , utilizing sedation with a ketamine/xylazine combination and euthanasia by intracardiac potassium chloride [13] . Rabies suspect animals that were euthanized or found dead were tested at the MARNDR National Veterinary Laboratory . Investigation results were reported by telephone to the bite victim , the reporting health facility , and the MSPP–Department of Epidemiology and Laboratory Resources . Additional persons identified as exposed during the course of community bite investigations were referred to nearby medical facilities for rabies post-exposure prophylaxis ( PEP ) evaluation . HARSP investigation forms collected information on the overall health status of the animals , presenting clinical signs , health history , and human exposures . Monitoring of adherence to rabies PEP is not collected in Haiti nor was it collected by HARSP . Active surveillance was conducted through convenience sampling of found-dead dogs from June 2013 through December 2014 . During this time , US embassy staff , during routine driving activities in the West Department , retrieved found-dead dogs for rabies testing . This type of sampling is assumed to be less biased compared to post-bite animal rabies surveillance , since it does not rely upon symptomatic indicators for investigation [8] . A case definition was developed to assign a case status to animals investigated through HARSP ( Box 1 ) . Investigation and diagnostic data collected from January 2013 through December 2014 were compared with historical records of animal rabies cases collected by MARNDR dating back to September 2009 . Data were entered into a Microsoft Access database and exported to SAS ( version 9 . 3 , SAS Institute Inc . , Cary , NC , USA ) . Univariate , descriptive analysis for temporal and spatial trends were conducted . Odds ratio ( OR ) and conditional maximum likelihood estimate test of association between clinical signs and case definition ( i . e . confirmed case , probable case , suspect case , or non-case ) were examined to validate the case definitions and determine factors significantly associated with rabid animals . A formal waiver was obtained from the National Center for Emerging Zoonotic Infectious Diseases human subject’s advisor; this work was deemed exempt , non-research .
During the two-year period in which HARSP operated , 70 confirmed rabid animals , 36 probable , 178 suspect , and 494 non-cases were reported ( Table 1 ) . The HARSP registered 738 passive and 40 active surveillance investigations ( Fig 2 ) . Of these 778 investigations , 143 animals ( 18 . 0% ) were tested for rabies and 70 ( 9 . 0% ) were confirmed positive; 66 through passive surveillance and 4 through active surveillance . Of confirmed rabid animals , 62 were dogs , 4 were cats , and 4 were goats . Probable rabid animals included 33 dogs , 2 goats , and 1 pig . Overall , passive and active surveillance streams contributed 106 confirmed and probable animals ( 13 . 6% of all animals reported to HARSP for assessment ) . Prior to the development of HARSP , from September 2009 through December 2012 , there were a total of 12 rabid animals reported in all of Haiti; an average of 0 . 3 rabid animals per month ( Fig 3 ) . For the first six months in which HARSP operated , 13 confirmed and probable rabies cases were reported ( average of 2 . 2 per month ) , 29 during the following six months ( average of 4 . 8 per month ) , and 64 during 2014 ( average of 5 . 3 per month ) . The reporting rate in 2014 represented an 18-fold increase in monthly reported rabid animals compared to pre-HARSP surveillance capacity . HARSP investigations were conducted in 28 ( 20% ) of Haiti’s 140 communes , of which rabid animals were found in 22 ( 78 . 6% ) ( Fig 4 ) . The six communes without detection of a rabid animal had fewer than six investigations conducted during the two-year period . The three primary HARSP communes accounted for 507 investigations ( 68 . 5% ) , and resulted in the recognition of 56 ( 52 . 8% ) of the 106 confirmed and probable cases . Confirmed and probable cases made up 14 . 5% of investigations in Petionville , 13 . 9% in Croix-des-Bouquets , and 6 . 0% in Carrefour . An average of 1 . 7 investigations were conducted in communes where no rabid animal had been identified , compared to an average of 36 . 2 investigations in communes with rabid animals . Among all communes , a confirmed or probable rabid animal was identified for every 7 . 4 investigations conducted . Signs of rabies illness were recorded as part of the HARSP investigation . The most common signs of illness among confirmed and probable animal cases were unusual aggression and biting , however these signs were also common among suspect and non-cases ( Table 2 ) . Confirmed and probable cases were significantly more likely to show signs of hypersalivation ( OR = 50 . 3 and 9 . 6 ) , paralysis ( OR = 6 . 7 and 3 . 6 ) , and lethargy ( OR = 9 . 0 and 7 . 3 ) compared to non-cases . There were no significant differences in signs of illness among suspect animals compared to non-cases . Twenty-four confirmed and probable rabid dogs died or were euthanized during the observation period , of which the average death occurred 3 . 0 days after observation was initiated ( range 0–9 days ) . A total of 639 bite victims were reported to HARSP and an additional 364 bite victims were identified during the course of investigations ( Table 3 ) . Of the total bite victims ( n = 1 , 003 ) , 137 ( 13 . 7% ) were likely exposed to rabies ( i . e . from both confirmed and probable cases ) and 48 ( 35% ) of these bite victims were identified as a result of the HARSP investigation . Only 43 ( 31 . 4% ) people with likely rabies exposures had initiated rabies PEP at the time of the HARSP investigation; the remaining 94 persons identified as exposed were referred by the ARSOs to medical centers for further evaluation by a medical provider . While not specifically collected by HARSP , no known human rabies cases emerged from persons identified through HARSP . Two human rabies cases were detected during HARSP community-based investigation activities . In both instances the human case had died several weeks preceding HARSP activities in the community and neither cases were reported to health officials prior to HARSP detection .
The rabies surveillance system developed in Haiti is based on standard surveillance practices applied to many human and animal diseases: case identification , contact tracing , epidemiologic investigation , and laboratory confirmation [14] . These are also the main principles put forth in the canine rabies blueprint , which supports that reliable and routine animal rabies diagnostic and surveillance capacity is the foundation upon which successful rabies control programs are established [10] . While the canine blueprint recommendations are readily available , it may be difficult for low-resource countries to implement all of these broad recommendations within the infrastructural confines of a developing economy . The implementation of HARSP was unique in that it was developed to operate within the infrastructural confines of both the ministries of health and agriculture , utilizing existing systems that were enhanced through the provision of targeted trainings and appropriate equipment . The HARSP concept was originally created in 2011 as a platform in which the following targets were attempted simultaneously: diagnostics , animal surveillance , human surveillance , education , canine vaccination , and population management . These ambitious goals have been successful in developed countries where resources for such activities are readily available [6] . However , in what may be consistent with many developing countries , this multi-pronged , simultaneous approach to rabies control proved too resource-intensive for the existing infrastructure in Haiti both in terms of human and monetary capacity . In late 2012 the HARSP concept was re-focused on two main targets: diagnostic development , followed by establishment of a routine animal rabies surveillance system in three pilot communes . More resource intensive objectives , which often rely on baseline epidemiologic and surveillance acquired knowledge , were delayed until these foundational programs were established . Establishing reliable case definitions is an integral component of any surveillance program . Animals that test positive or negative and animals that pass an observation period fall into standard case definitions: confirmed and non-case . However , animals that are not located , and those lacking diagnostic confirmation are more difficult to classify within standard case definitions . Among the four HARSP case definitions ( confirmed , probable , suspect , and non-cases ) there was no interpretable difference among the clinical signs of biting and aggression . This was expected , as these signs were the impetus for most HARSP investigations . Clinical illness consisting of hypersalivation , paralysis , and lethargy occurred significantly more in confirmed and probable animals than in non-cases . It is unlikely that all probable cases had rabies , but these strong associations may indicate that many animals within the probable case definition category had the disease . Clinical signs reported among suspect rabid animals were very similar to signs reported in non-cases; perhaps an indication that rabies was unlikely within the suspect case definition group . However , suspect animals are those which were not located or assessed by ARSOs . Therefore , the clinical descriptions were based upon bite victim or community member reports , and not through the assessment of trained professionals . It is possible that a proportion of these suspect animals would have fallen into confirmed or probable case definitions had they been located and properly assessed . Overall , it would appear that the case definitions chosen for this rabies surveillance program are appropriate and should be considered for other similar surveillance programs which are trying to further the understanding of the epizootiology of endemic canine rabies . The HARSP detected 106 confirmed and probable rabies cases in the first two years while primarily focusing efforts in just three of Haiti’s 140 communes . This two-year count represented a 10-fold increase in detection of rabid animals compared to the previous two year’s reports for the entire country , further supporting the utility of this type of surveillance model . The program appeared to improve detection capacity as time progressed , with a 140% increase in monthly case detection during the last year of the program compared to when it began . This is likely a reflection on the continued training and proficiency of ARSOs and increased community and medical facility awareness of the program . With such low rabies detection levels prior to implementation of the HARSP , and numerous competing health problems , rabies was not among the countries top health priorities . Yet , published estimation methods have predicted that there are likely 130 or more human deaths , indicating a substantial underreporting of rabies in the country [2] . The information collected through HARSP provides the first systematically collected data indicating that the animal ( and likely human ) rabies burden is much higher than currently recognized . The finding that 9% of dogs assessed through passive means and 10% of dogs assessed through active means were confirmed rabid is particularly concerning . Few canine endemic countries have published the outcomes of post-bite animal rabies surveillance investigations , however several studies have published results of the rabies test percent positive during similar activities . In Haiti , rabies virus was confirmed in 66% of suspect animals which were tested . Highly endemic canine rabies affected countries have reported rates similar to or lower than what has been observed in Haiti . A study in Kenya reported 51% positive test rate among suspect cases tested , 68% in Tanzania , and 71 . 7% in Bhutan [3 , 15 , 16] . A roadside surveillance program in Kenya found 15% of found-dead dogs were positive for rabies virus , a figure comparable with what has was recorded through the HARSP [3] . These countries , Kenya , Tanzania , and Bhutan , have reported human rabies death rates ranging from 4 . 7–19 . 2 per 100 , 000 population [15 , 16] . This study has confirmed a level of enzootic canine rabies activity in Haiti similar to that found in countries with high human rabies death rates; a finding that warrants further examination of the potential human rabies burden . If Haiti is afflicted with human rabies death rates in these comparable canine endemic countries , several hundred undocumented deaths may occur each year . Despite HARSP being theatrically located in three communes , its actions extended to neighboring communes , where rabid animals were found in every commune where at least seven investigations were conducted . Only an estimated 8 . 5% of Haitians reside in the three primary HARSP communes . These communes were not chosen because of preconceived concerns about elevated rabies activity; rather they were chosen based on proximity to the national laboratory and existing infrastructure such as accessible roads and hospitals . The three communes represented both high and low socioeconomic areas , yet the highest case rate was from the wealthiest commune , Petionville ( 14 . 5% ) . The three HARSP communes also may have higher rates of canine rabies vaccination coverage compared to many parts of the country , as vaccination campaigns have been conducted in these communities during three of the last five years ( verbal communication , MARNDR ) . Lastly , these communes are urban , whereas numerous studies have shown that canine rabies vaccination rates are lower and access to PEP is more difficult in rural settings [17] . Routine rabies surveillance has never been conducted in a majority of Haitian communes , including several large urban centers . Further expansion of the HARSP to serve more communes will improve the epidemiologic understanding of this disease and assist in prevention efforts . Passive surveillance is dependent upon reports from medical center and community channels . When these communication channels are functional , accurate extrapolation of results may be possible . The human population of the primary surveillance area was: 890 , 000 . Hampson et al have estimated annual bite rates of 239/100 , 000 in developing countries such as Haiti [2] . Applying this rate to the population of the three HARSP communes , during the two-year time period we would have expected approximately 4 , 254 bites to have occurred in these communities . Only 738 ( 17% ) of the estimated total bites were reported to rabies officers . This has two potential impacts on interpretation of this data . With such low reporting of bite events , there may have been more rabies activity in the community that went unrecognized . If this is the case , then the true number of canine and human rabies cases is much higher in the primary HARSP communes than currently understood . Alternatively , the event of a rabid animal in a community may be more easily recognized and reported . Under this scenario , the large percentage of bites that are not reported may be trivial bites by healthy , known dogs , in which case these surveillance data would not merit extrapolation . Increased coverage of HARSP investigations in the three primary communes could help to further the understanding of true rabies burden . Rabies is one of the world’s most feared diseases and is responsible for more deaths globally than any other zoonotic disease [1] . Yet , rabies is also a readily preventable disease with effective human and animal biologics , effective population management strategies , and effective surveillance programs that have been developed and proven economically efficient [4 , 10 , 20] . Unfortunately , rabies remains a neglected disease in many countries , in large part due to the cycle of neglect that starts with poor epidemiologic and epizootiologic information about the disease . Haiti is one of only four countries in the western hemisphere that still records human deaths due to canine rabies [6] . The HARSP has provided some of the first measured estimates of rabies in Haiti , and has shown that the burden and distribution of this disease is large and likely widespread . The HARSP has also shown that it can prevent human rabies deaths through active case investigations and community outreach . Programs such as HARSP are the foundation upon which successful rabies control strategies are built . The development of surveillance programs employing concepts of HARSP should be considered in countries where canine rabies remains endemic . | The Republic of Haiti has highest estimated burden of human rabies in the Western Hemisphere , at 130 estimated human deaths annually . Rabies surveillance systems in the majority of the developing world , including Haiti , are ineffective , resulting in underreporting of cases and contributing to the further neglect of this disease . In 2013 a passive rabies surveillance program was implemented in three of Haiti’s 140 communes near the nation’s capital city . Four animal rabies surveillance officers conducted 778 suspect animal rabies investigations in a two-year period and on average found a rabid animal for every 7 . 4 investigations . Prior to the implementation of this surveillance program Haiti reported an average of two canine and seven human rabies cases each year , for the entire country . This program identified 70 rabid animals and an additional 36 probable rabid animals in only a selected area of the country . These 106 cases represent an 18-fold increase in animal rabies reporting in Haiti . These findings support that canine rabies is a significant burden in Haiti and present data that can be used to improve human rabies burden estimations and enhance canine rabies control efforts . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Establishment of a Canine Rabies Burden in Haiti through the Implementation of a Novel Surveillance Program |
The sexually transmitted bacterium Neisseria gonorrhoeae has developed resistance to all antibiotic classes that have been used for treatment and strains resistant to multiple antibiotic classes have evolved . In many countries , there is only one antibiotic remaining for empirical N . gonorrhoeae treatment , and antibiotic management to counteract resistance spread is urgently needed . Understanding dynamics and drivers of resistance spread can provide an improved rationale for antibiotic management . In our study , we first used antibiotic resistance surveillance data to estimate the rates at which antibiotic-resistant N . gonorrhoeae spread in two host populations , heterosexual men ( HetM ) and men who have sex with men ( MSM ) . We found higher rates of spread for MSM ( 0 . 86 to 2 . 38 y−1 , mean doubling time: 6 months ) compared to HetM ( 0 . 24 to 0 . 86 y−1 , mean doubling time: 16 months ) . We then developed a dynamic transmission model to reproduce the observed dynamics of N . gonorrhoeae transmission in populations of heterosexual men and women ( HMW ) and MSM . We parameterized the model using sexual behavior data and calibrated it to N . gonorrhoeae prevalence and incidence data . In the model , antibiotic-resistant N . gonorrhoeae spread with a median rate of 0 . 88 y−1 in HMW and 3 . 12 y−1 in MSM . These rates correspond to median doubling times of 9 ( HMW ) and 3 ( MSM ) months . Assuming no fitness costs , the model shows the difference in the host population’s treatment rate rather than the difference in the number of sexual partners explains the differential spread of resistance . As higher treatment rates result in faster spread of antibiotic resistance , treatment recommendations for N . gonorrhoeae should carefully balance prevention of infection and avoidance of resistance spread .
Antibiotic-resistant Neisseria gonorrhoeae can evolve and spread rapidly [1] . Resistance is commonly observed against the antibiotic classes penicillin , tetracycline and fluoroquinolones [2–4] . Resistance also emerged against cefixime , an oral third generation cephalosporin , in recent years [2 , 3] . Since 2010 , cefixime is no longer recommended as first-line treatment [5] following guidelines from the World Health Organization ( WHO ) that an antibiotic should not be used when more than 5% of N . gonorrhoeae isolates are resistant [6] . Injectable ceftriaxone , in combination with oral azithromycin , is now the last antibiotic remaining as recommended first-line treatment [7] . Although other antibiotics are being tested for their safety and efficacy for N . gonorrhoeae treatment [8] , no new classes of antibiotics are currently available [4] and management of antibiotics is urgently needed to preserve their efficacy . The current management strategy tries to reduce the overall burden of N . gonorrhoeae infection by expanded screening and treatment of hosts [9 , 10] , but the outcome of this strategy for resistance is uncertain . Understanding the drivers of resistance spread and anticipating future resistance trends will provide rationales for antibiotic management and help to improve antibiotic treatment strategies . Men who have sex with men ( MSM ) are host populations that have higher levels of antibiotic-resistant N . gonorrhoeae than heterosexual host populations [3] . In a study [5] based on the Gonococcal Resistance to Antimicrobials Surveillance Programme ( GRASP ) in England and Wales , cefixime-resistant N . gonorrhoeae were mainly found in MSM until 2011 . The authors suggested that cefixime resistance was circulating in a distinct sexual network of highly active MSM and that bridging between MSM and heterosexuals was necessary for subsequent spread among heterosexual hosts . However , cefixime-resistant N . gonorrhoeae might have already been spreading undetected in the heterosexual host population . Mathematical models can help explain the differential observations of antibiotic-resistant N . gonorrhoeae in different host populations . In 1978 , Yorke et al . [11] introduced the concept of core groups to model the transmission of N . gonorrhoeae . The concept of core groups posits that an infection can only be maintained in a host population if a highly sexually active group of hosts is responsible for a disproportionate amount of transmissions . More recent modeling studies have examined the transmission of antibiotic-resistant N . gonorrhoeae . Chan et al . [12] found that prevalence rebounds more quickly to a pre-treatment baseline when treatment is focused on the core group . Xiridou et al . [13] developed a N . gonorrhoeae transmission model to determine the impact of different treatment strategies on the prevalence of N . gonorrhoeae in Dutch MSM . They found that increased treatment rates could increase the spread of resistance , whereas re-treatment could slow it down . Hui et al . [14] used an individual-based N . gonorrhoeae transmission model in a heterosexual host population to investigate the effect of a molecular resistance test on the time until 5% resistance are reported . None of these studies has investigated or explained the differences in the spread of antibiotic-resistant N . gonorrhoeae in MSM and heterosexual host populations . In this study , we investigated the dynamics and determinants of antibiotic-resistant N . gonorrhoeae spread using surveillance data and mathematical modeling . We estimated the rates at which resistance spreads in heterosexual men ( HetM ) and MSM using surveillance data from the USA and from England and Wales . We then developed a mathematical model of N . gonorrhoeae transmission to reconstruct the observed dynamics of resistance spread . This allowed us to determine the major driver of resistance spread , and to explore the expected rates at which resistance spreads in MSM and heterosexual host populations .
We fitted a logistic growth model to the proportion of antibiotic-resistant N . gonorrhoeae as observed in the two gonococcal surveillance programs ( Fig 2 ) . The proportion of cefixime-resistant N . gonorrhoeae in GRASP appears to increase for both HetM and MSM after 2006 . Ciprofloxacin-resistant N . gonorrhoeae in HetM and MSM were spreading in all observed host populations after the year 2000 . For a given antibiotic and surveillance program , the rates of resistance spread were consistently higher for MSM than for HetM ( Table 4 ) . The average rate of resistance spread was 0 . 53 y−1 for HetM and 1 . 46 y−1 for MSM , corresponding to doubling times of 1 . 3 y ( HetM ) and 0 . 5 y ( MSM ) during the initial exponential growth phase . Next , we studied the transmission of N . gonorrhoeae and the spread of resistance in the dynamic transmission model . We calibrated five model parameters to expected prevalence and incidence in MSM and HMW host populations . The posterior distributions of the parameters were based on 2 , 779 parameter sets for HMW and 65 , 699 parameter sets for MSM ( Fig 3 , Table 1 ) . Distributions of the modeled prevalence and incidence of diagnosed infections after calibration are provided as Supporting Information ( S1 and S2 Figs , S3 Table ) . The sexual mixing coefficient showed a tendency towards assortative mixing in both MSM and HMW ( Fig 3a ) . The fraction of diagnosed and treated infections tended to be higher in MSM compared to HMW ( Fig 3b ) , whereas the infectious duration was considerably shorter in MSM ( median: 2 . 3 months , IQR: 1 . 7–3 . 0 months ) than in HMW ( median: 6 . 6 months , IQR: 5 . 5–7 . 9 months ) ( Fig 3c ) . The transmission probabilities per partnership were generally higher in HMW than in MSM ( Fig 3d and 3e ) . After calibration , we used the dynamic transmission model to study the spread of antibiotic-resistant N . gonorrhoeae . The proportion of antibiotic-resistant N . gonorrhoeae increased faster in MSM than in HMW ( Fig 4 ) . In HMW , the median of all simulations reached 5% resistance in fewer than 4 . 5 y and 50% resistance in fewer than 7 . 8 y after appearance of the first antibiotic-resistant N . gonorrhoeae infection . In the MSM population , the median of all simulations reached a resistance level of 5% in fewer than 1 . 7 y and 50% in fewer than 2 . 6 y after resistance first appears in the population . The range spanned by all simulations was much wider in HMW than in MSM: 95% of all simulations reached the 5% threshold in fewer than 2 . 7–7 . 7 y ( HMW ) , compared with 1 . 1–2 . 2 y ( MSM ) . Antibiotic-sensitive and -resistant N . gonorrhoeae share the same resource for growth , i . e . the susceptible hosts . The rate at which one strain replaces the other strain in the host population is given by the difference in their net growth rates . We assume that the transmission probabilities and the infectious duration of the two strains are the same . Since the probability of resistance during treatment is very small ( μ ≪ 1 ) , the difference in net growth rates between the strains is approximated by the treatment rate τ and corresponds to the rate of spread of antibiotic-resistant N . gonorrhoeae . The observed distributions of treatment rates from the transmission model hardly overlap between HMW and MSM ( Fig 5 ) . The median treatment rates , i . e . the approximated median rates of resistance spread in the transmission model are 3 . 12 y−1 ( MSM ) and 0 . 88 y−1 ( HMW ) . We tested whether changes in the probability of resistance during treatment , μ , and fitness costs in the antibiotic-resistant strain alter the model outcomes . Higher probabilities of resistance during treatment accelerate the establishment of antibiotic-resistant N . gonorrhoeae in the population and hence reduce the time until 5% resistance is reached ( S3 Fig ) . Higher probabilities of resistance during treatment , however , do not affect rates of spread , unless the probability of resistance during treatment is unrealistically high ( 10% ) ( S4 Fig ) . Fitness costs in the antibiotic-resistant strain result in rates of resistance spread that are lower than the treatment rate τ ( Fig . B in S1 Appendix ) . Fitness costs that reduce the transmission probability per partnership , βij , have a stronger effect than fitness costs that reduce the duration of infection . The effects of fitness costs are independent of the sexual partner change rate , πi , and βij if they affect the duration of infection , but can vary with πi and βij if they affect the transmission probability per partnership ( Fig . C in S1 Appendix ) . While high fitness costs can prevent the spread of antibiotic-resistant strains ( Fig . A in S1 Appendix ) , fitness costs between 0%–10% have only small effects on the rates of resistance spread ( Fig . B in S1 Appendix ) .
In this study , we quantified the rate at which antibiotic-resistant N . gonorrhoeae spread in heterosexual and MSM populations . We used data from two different surveillance programs and estimated that the proportion of ciprofloxacin- and cefixime-resistant N . gonorrhoeae doubles on average every 1 . 3 y in HetM and 0 . 5 y in MSM . The faster spread of antibiotic-resistant N . gonorrhoeae in MSM than in heterosexual hosts was corroborated using a dynamic transmission model , which was calibrated to observed prevalence and incidence rates . The model allowed us to identify the higher treatment rates in MSM , compared with heterosexual hosts , as the major driver for the faster spread of antibiotic-resistant N . gonorrhoeae . To our knowledge , this is the first study to have analyzed and interpreted N . gonorrhoeae antibiotic resistance surveillance data in a dynamic and quantitative manner . The transmission model was parameterized using sexual behavior data for HMW and MSM from Natsal-2 [22] , a large probability sample survey of sexual behavior . Calibrating the model to observed prevalence and incidence rates allowed us to use largely uninformative priors for the model parameters . The calibration makes our model more robust to changes in parameters than using fixed parameter values , especially since for N . gonorrhoeae available parameter values are very uncertain [31] . It also allowed us to rely on few assumptions about the natural history of N . gonorrhoeae infection . The limitations to our study need to be taken into consideration when interpreting the findings . First , we used data from different sources , although all were collected in high income countries . The antibiotic resistance surveillance data are from programs in England and Wales and the USA . The mathematical transmission model was parameterized using British sexual behavior data [22] and calibrated to prevalence and incidence rates from the USA ( HMW ) [26 , 27] and Australia ( MSM ) [28 , 29] . For simplicity , we modeled the heterosexual and MSM host populations separately although there is some mixing between them . We assumed the sexual behavior of heterosexual men and women to be the same and pooled their behavioral data . Second , we assumed complete resistance against the antibiotic , i . e . 100% treatment failure . We further assumed that treatment of the sensitive strain is 100% efficacious . Both assumptions might explain why antibiotic-resistant N . gonorrhoeae spread at somewhat higher rates in the dynamic transmission model than estimated from data . Third , we restricted our model to resistance to one antibiotic with no alternative treatment or interventions . This is why we observe complete replacement of the antibiotic-sensitive strain in the model , a phenomenon that has not been observed in surveillance data . Fourth , resistance in our model is treated as a generic trait , but it likely depends on the underlying molecular mechanisms and possibly the genetic background of the N . gonorrhoeae strain . Different resistance mechanisms might explain some of the differences in the rates of resistance spread between the model and the different antibiotics from the surveillance data . Fifth , we did not include co- and superinfection with antibiotic-sensitive and -resistant N . gonorrhoeae strains . Since genetic typing provides evidence for mixed infections [32] , it is worth speculating how they would affect the rate of spread from the transmission model . If antibiotic-sensitive and -resistant strains co-existed in a host and acted independently , we would not expect significant effects on the rate of spread . In contrast , if there was competition between the two strains within a host , the rate of spread would increase if the antibiotic-resistant strain outcompetes the -sensitive strain , and decrease otherwise . Sixth , we do not consider importation of resistance from another population . For example , importation of resistance from other countries might play a particularly important role during the early phase of resistance spread , when stochastic events can lead to extinction of the antibiotic-resistant strain . We expect that a high rate of importation of antibiotic resistance shortens the time to reach 5% resistance drastically , but that once the resistant strain is established in the population , importation hardly affects the rate of resistance spread . Finally , we assumed that the transmission probabilities per partnership and the durations of infection in the model represent average values for N . gonorrhoeae infections at different infection sites ( urethral , pharyngeal , anal , cervical ) . The estimated posterior distributions of the parameters fit within the range of previously used values , and provide some insights into sexual mixing and the natural history of N . gonorrhoeae . The sexual mixing coefficient tends to be assortative for both HMW and MSM host populations in our model . Quantifying the degree of sexual mixing is difficult and largely depends on the study population , but our finding is consistent with other studies indicating assortative sexual mixing in the general population [30 , 33] . The posterior estimates of the fraction of diagnosed and treated infections are consistent with the notion that a large proportion of N . gonorrhoeae infections are symptomatic , and that this proportion is expected to be higher in men than in women [34–36] . The average duration of infection was the only parameter with an informative prior , but we found marked differences between the duration of infection in HMW ( 6 . 6 months ) and MSM ( 2 . 3 months ) . Per sex act transmission probabilities are generally considered to be lower from women to men than vice versa [37–39] . In our model , the median of the transmission probability per partnership was lower in MSM hosts than in HMW for both sexual activity groups . This could be explained by different numbers of sex acts per partnership in the two populations . The low transmission probability within the highly active MSM group ( median: 30% ) could reflect a single or a small number of sex acts per partnership . In contrast , the high transmission probability for HMW within the low sexual activity group ( median: 87% ) could be a result of a larger number of sex acts per partnership in those individuals . Furthermore , condom use is more frequent in MSM than in HMW [22] , which could explain part of the observed differences in transmission probabilities . Our study found that the treatment rate is the driving force of resistance spread . Xiridou et al . [13] found that resistance could spread faster when the treatment rate was higher , but they did not identify the treatment as the major driver of resistance spread . Chan et al . [12] found that focusing treatment on the core group leads to a faster rebound to pre-treatment prevalence than equal treatment of the entire host population . Unfortunately , our findings cannot be compared with Chan et al . because they do not report the proportion of antibiotic-resistant N . gonorrhoeae . It was shown previously that treatment is the main selective force acting on resistance evolution due to the selective advantage to the resistant pathogen [40 , 41] . We now expand this concept by showing that , assuming no fitness costs , treatment rates determine the rates of resistance spread even when the host populations has a heterogeneous contact structure . The intuitive argument that a faster spread of an infection , due to a higher number of sexual partners , will result in a faster spread of resistance does not hold . Instead , the proportion of resistant infections spreads equally in host populations with different number of partners as long as they receive treatment at the same rate and there are no fitness costs associated with the transmission probability per partnership . For N . gonorrhoeae , this insight challenges the current management strategy that aims to lower the overall burden of infection by expanding screening and treatment of hosts [9 , 10] . As soon as antibiotic-resistant pathogens are frequent enough to evade stochastic extinction , expanded treatment will foster their spread and increase the burden of N . gonorrhoeae . Additionally , we show that fitness costs can decelerate or even prevent the spread of antibiotic-resistant N . gonorrhoeae strains . Fitness costs therefore might explain why highly resistant strains , such as the ceftriaxone-resistant N . gonorrhoeae strain H041 , do not spread in the host population after their first detection [42] . Our findings also show that bridging between the HetM and the MSM host populations might not have been necessary for cefixime-resistance to spread in the HetM population after 2010 [5] . It is likely that cefixime-resistant N . gonorrhoeae had already been present in the HetM population but were spreading at a lower rate than in the MSM population . The results of our study will be useful for future N . gonorrhoeae research and for guiding treatment recommendations . The N . gonorrhoeae transmission model describes observed prevalence and incidence rates well and can reconstruct the spread of antibiotic-resistant N . gonorrhoeae . Estimating rates of resistance spread is useful for projecting future resistance levels and the expected time it will take until a certain threshold in the proportion of antibiotic-resistant N . gonorrhoeae is reached . Until now , treatment recommendations for N . gonorrhoeae are subject to change when 5% of N . gonorrhoeae isolates show resistance against a given antibiotic [6] . Our study shows the importance of the rate of spread: a level of 5% resistance results in a marginal increase to 8% in the following year if resistance spreads logistically at rate 0 . 53 y−1 ( HetM mean estimate from Table 4 ) , but reaches 18% in the next year if resistance spreads at rate 1 . 46 y−1 ( MSM mean estimate from Table 4 ) . Public health authorities could use surveillance data and adapt thresholds for treatment recommendation change to specific host populations using the method we describe . Our study challenges the currently prevailing notion that more screening and treatment will limit the spread of N . gonorrhoeae , as higher treatment rates will ultimately result in faster spread of antibiotic resistance . Future treatment recommendations for N . gonorrhoeae should carefully balance prevention of N . gonorrhoeae infection and avoidance of the spread of resistance . | More and more infectious disease treatments fail because the causative pathogens are resistant to the drugs used for treatment . For the treatment of Neisseria gonorrhoeae , a sexually transmitted bacterium , drug resistance is a particularly big problem: there is only a single antibiotic left that is recommended for treatment . We aimed to understand how antibiotic-resistant N . gonorrhoeae spread in a sexually active host population and how the spread of resistance can be slowed . From antibiotic resistance surveillance data , we first estimated the rate at which antibiotic-resistant N . gonorrhoeae spread . Second , we reproduced the observed dynamics in a mathematical model describing the transmission between hosts . We found that antibiotic-resistant N . gonorrhoeae spread faster in host populations of men who have sex with men than in host populations of heterosexuals . We could attribute the faster spread of resistant pathogens to higher treatment rates . This finding implies that promoting screening to control antibiotic-resistant N . gonorrhoeae could in fact accelerate their spread . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
] | [
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"drugs",
"microbiology",
"neisseria",
"gonorrhoeae",
"antibiotic",
"resistance",
"probability",
"distribution",
"mathematics",
"antibiotics",
"sexually",
"t... | 2016 | Antibiotic-Resistant Neisseria gonorrhoeae Spread Faster with More Treatment, Not More Sexual Partners |
The circulation of vector-borne zoonotic viruses is largely determined by the overlap in the geographical distributions of virus-competent vectors and reservoir hosts . What is less clear are the factors influencing the distribution of virus-specific lineages . Japanese encephalitis virus ( JEV ) is the most important etiologic agent of epidemic encephalitis worldwide , and is primarily maintained between vertebrate reservoir hosts ( avian and swine ) and culicine mosquitoes . There are five genotypes of JEV: GI-V . In recent years , GI has displaced GIII as the dominant JEV genotype and GV has re-emerged after almost 60 years of undetected virus circulation . JEV is found throughout most of Asia , extending from maritime Siberia in the north to Australia in the south , and as far as Pakistan to the west and Saipan to the east . Transmission of JEV in temperate zones is epidemic with the majority of cases occurring in summer months , while transmission in tropical zones is endemic and occurs year-round at lower rates . To test the hypothesis that viruses circulating in these two geographical zones are genetically distinct , we applied Bayesian phylogeographic , categorical data analysis and phylogeny-trait association test techniques to the largest JEV dataset compiled to date , representing the envelope ( E ) gene of 487 isolates collected from 12 countries over 75 years . We demonstrated that GIII and the recently emerged GI-b are temperate genotypes likely maintained year-round in northern latitudes , while GI-a and GII are tropical genotypes likely maintained primarily through mosquito-avian and mosquito-swine transmission cycles . This study represents a new paradigm directly linking viral molecular evolution and climate .
Japanese encephalitis virus ( JEV ) belongs to the JEV serocomplex within the genus Flavivirus , family Flaviviridae . Recurrent epidemics of summer encephalitis suggestive of JE were recorded in Japan from 1871 onwards and major epidemics occurred in 1924 ( 6 , 000 cases , with a 60% case fatality rate ) , 1929 , 1935 and 1937 [1] . The prototype Nakayama strain of JEV was isolated in mice from the brain of a male that died of summer encephalitis in Tokyo , Japan in 1935 [1] . The seasonal occurrence of epidemic encephalitis coupled with the abundance of culicine mosquitoes led to suggestions that JEV was transmitted by mosquito vectors , leading to the subsequent recovery of the virus from rice-paddy breeding Culex tritaeniorhynchus mosquitoes in 1938 [2] . A series of ecological studies performed in Japan in the late 1950s established waterbirds as maintenance hosts of the virus , domestic swine as major amplifying hosts , and Cx . tritaeniorhynchus as the principal vector between these vertebrate hosts and the incidental , dead-end human host [3] , [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] . JEV circulates throughout most of Asia , with the northern limit of virus activity extending north into maritime Siberia . In recent years the geographical distribution of JEV has expanded , reaching east into Saipan in 1990 [13] , west into Pakistan in 1992 [14] and south into the Torres Strait between Papua New Guinea and Australia in 1995 [15] . JEV epidemics occur in temperate zones , with the majority of cases occurring in summer or monsoon season months . In contrast , JEV is endemic in tropical regions and transmission occurs year-round at lower rates [16] . Despite the availability of effective vaccines against JEV , the virus is still considered the most important etiologic agent of epidemic encephalitis worldwide , causing an estimated 68 , 000 cases and a reported 10 , 000–15 , 000 deaths annually [17] . Of symptomatic infections , 20–30% are rapidly fatal , 30–50% develop long-term neurologic and/or psychiatric sequelae , and only 20–50% fully resolve the disease [17] . Like other flaviviruses , JEV possesses an 11 kilobase , single-stranded , positive-sense RNA genome containing 5′ and 3′ untranslated regions , and a single open reading frame ( ORF ) encoding a polyprotein that is co- and post-translationally cleaved by viral and host proteases into three structural proteins: the capsid ( C ) , the precursor of the membrane ( prM ) , and the envelope ( E ) , as well as seven non-structural proteins [18] . The E protein represents the major constituent of the mature virion surface and is the dominant antigen involved in the elicitation of virus neutralizing antibodies [19] . Phylogenetic studies have divided JEV into five genotypes . GI includes isolates collected in northern Australia , northern Cambodia , China , India , Japan , Korea , Laos , Malaysia , Taiwan , Thailand and Vietnam between 1967 and present . GII includes isolates collected sporadically in northern Australia , Indonesia , Korea , Malaysia , Papua New Guinea and southern Thailand between 1951 and 1999 . GIII has been the source of annually occurring epidemics of encephalitis and includes isolates collected in China , India , Indonesia , Japan , Korea , Malaysia , Myanmar , Nepal , Philippines , Sri Lanka , the former Soviet Union , Taiwan , Thailand and Vietnam between 1935 and present . GIV includes seven isolates collected in Indonesia between 1980 and 1981 from mosquitoes only . GV includes three isolates collected in Malaysia , China and South Korea between 1952 and 2010 . Previous investigations noted GI and GIII viruses were collected mostly in temperate zones , while GII and GIV isolates were collected mostly in tropical zones [20] , [21] . However , the statistical significance of this observation has never been tested using a comprehensive dataset of JEV isolates , with information regarding the isolates' genotypes and locations of collection . The geographical distribution of JEV has expanded in recent years , causing outbreaks of encephalitis in immunologically naïve populations . In addition , the molecular epidemiology of the virus has changed over this period of time . From the isolation of the prototype Nakayama strain of JEV in 1935 until recently , GIII was the most frequently isolated genotype throughout Asia . However , over the past two decades , multiple reports have indicated that GI has displaced GIII as the most frequently isolated virus genotype in a number of Asian countries including China [22] , Thailand [23] , South Korea [24] , Japan [25] , Malaysia [26] , Vietnam [27] , India [28] and Taiwan [29] . Further , following the isolation of the GV Muar isolate [30] in 1952 from an encephalitic patient originating in Malaysia , the genotype remained undetected for almost 60 years until a pool of Cx . tritaeniorhynchus collected in the Tibetan Province of China in 2009 yielded the GV XZ0934 isolate [31] and a pool of Culex bitaeniorhynchus collected in South Korea in 2010 yielded the GV 10-1827 isolate [32] . A recent evolutionary study utilizing sequence information derived from the ORF of 35 JEV isolates ( 22 GIII isolates ) revealed that JEV originated from its ancestral virus around 1500 [33] and an earlier evolutionary study using 18 genomic JEV sequences ( 14 GIII ) proposed that this evolutionary event occurred in the Indonesia-Malaysia region [34] . Due to the small viral sequence sample sizes , neither of these studies were able to robustly examine the evolution , epidemiology or geographical distribution of the genotypes of JEV . In disagreement with the results of previous studies [33] , [34] , a recent study utilizing 98 genomic sequences , 76 of which were derived from Chinese virus isolates , estimated that JEV originated from its ancestral virus around 300 AD [35] . This difference was likely due to the slow evolutionary rate estimated for the Chinese JEV study [35] relative to previous JEV evolutionary studies [33] , [34] , as well other flavivirus evolutionary studies [36] , [37] , [38] , [39] . Prior to the work presented here , no studies have utilized a comprehensive dataset of molecular sequences to examine the phylogeography and epidemiology of the virus genotypes . Although there is a paucity of ORF sequences of wild-type isolates , extensive sequencing of the phylogenetically-informative E gene of both old and new JEV isolates in recent years has resulted in a large , spatiotemporally distributed collection of viral sequence data . Therefore , we performed a phylogeographic analysis on a dataset consisting of E gene sequence information derived from 487 JEV isolates ( largest collection of JEV sequences assembled to date ) to address the following key questions: 1 ) When and where did the virus and its genotypes originate , and what is their geographical range ? 2 ) Is there an association between genotype and climate of virus collection ( temperate versus tropical zones ) ? 3 ) What amino acid sites within the E protein were involved in the phylogenetic divergence of JEV and were any of these sites subject to diversifying and/or directional selection ?
All available sequences for the E gene of JEV isolates were retrieved from GenBank in July 2011 . The initial JEV E gene dataset was pruned of sequences representing non wild-type virus isolates , duplicate isolates , and isolates absent of information regarding the date and country of collection . The pruned dataset consisted of 489 sequences . The E gene sequences of two JEV isolates ( M859/Cambodia/1967/Mosquito and KE-93-83 ) obtained from the World Reference Center for Emerging Viruses and Arboviruses ( WRCEVA ) at the University of Texas Medical Branch ( UTMB ) , were determined for analysis in this study utilizing previously described methods [40] , [41] , [42] . Recombination can invalidate the results of coalescent analyses . Therefore , the nucleotide sequence alignment file was analyzed for potential recombination events using RDP [43] , GENECONV [44] , Chimaera [45] , MaxChi [46] and Bootscan [47] methods implemented in RDP3 v Beta 41 [48] . Common program settings were to perceive sequences as linear , require phylogenetic evidence , refine breakpoints and check alignment consistency , while all method-specific program settings remained at their default values . The highest acceptable p-value was set at 0 . 05 , after considering Bonferroni correction for multiple comparisons . Potential recombination events were those that were identified by at least two methods . The breakpoint positions and recombinant sequence inferred for the potential recombination events were manually confirmed using the phylogenetic and recombination signal analysis features in RDP3 . The K82P01 and K91P55 sequences were confirmed as recombinants ( Table S1 ) . These two isolates were not available from the WRCEVA at UTMB to re-sequence; therefore , the two corresponding sequences were removed from the dataset , leaving a final dataset of 487 sequences . To make an initial identification of the genotype of the JEV E gene sequences , neighbor-joining ( NJ ) and maximum-likelihood ( ML ) phylogenies were generated using SeaView v 4 . 2 . 12 [49] and PhyML v 3 . 0 on the South of France bioinformatics platform [50] , respectively . The final dataset of 487 JEV E gene sequences included information regarding the year , host and country of collection of the corresponding virus isolates . Sequences derived from isolates collected north of the Tropic of Cancer ( 23 . 5°N ) were classified as temperate , while sequences derived from isolates collected south of the Tropic of Cancer were classified as tropical . The climate corresponding to five Taiwanese sequences could not be ascertained and therefore these sequences were not included in the climate phylogeographic analysis described below . To estimate the date and location of the most recent common ancestor ( MRCA ) of the five genotypes and the overall rate of molecular evolution , time-scaled Bayesian phylogenies ( country and climate ) were inferred from the JEV E gene sequence dataset using a Bayesian Markov Chain Monte Carlo ( MCMC ) method implemented in BEAST v 1 . 6 . 1 [51] . An SDR06 nucleotide substitution model [52] , a relaxed-uncorrelated exponential molecular clock and a piecewise constant Bayesian skyline demographic model with 20 coalescent-interval groups [53] were used in all analyses . The relaxed-uncorrelated exponential molecular clock was found to best-fit the data when Bayes factor ( BF ) values were calculated ( Tracer v 1 . 5 . 1 ) [54] to evaluate the relative fit of strict and relaxed molecular clock models to the data by determining the natural logarithm of the ratio of the marginal likelihoods of the competing models [55] . The good fit of this relaxed clock model to the data has recently been shown to be an artifact of the harmonic mean estimator [56] . However , preliminary analyses showed that the selection of a particular relaxed molecular clock model had little effect on the results . To infer the probable geographic origin of the MRCA of the genotypes of JEV , the BEAST input files ( country and climate ) created in BEAUti v 1 . 6 . 1 [51] were edited to include the Bayesian stochastic search variable selection procedure [57] . The Bioportal at the University of Oslo [58] was used to execute the MCMC analyses for 600 million generations . This was achieved by using LogCombiner v 1 . 6 . 1 [51] to compile 12 independent runs of 50 million generations ( sampled every 1 , 000th state ) to attain convergence , which was assessed by examining the trace and effective sample size statistics for each model parameter in Tracer v 1 . 5 [54] . TreeAnnotator v 1 . 6 . 1 [51] was used to summarize the posterior tree distribution and annotate both country and climate maximum clade credibility ( MCC ) phylogenies , which were viewed in FigTree v 1 . 3 . 1 [59] . Each of the nodes of the Bayesian MCC phylogenies were annotated with posterior probability ( PP ) values , estimated median dates of the MRCA with corresponding 95% HPD values , and state PP values for each plausible geographic location of origin ( country and climate ) . In addition , BOA v 1 . 15 [60] implemented in R v 2 . 15 . 1 [61] was used to calculate a 50% HPD interval for the date of the root of the phylogeny . Maps showing the distributions of sequences according to sampling location ( country and climate ) were created using GIMP v 2 . 6 . 12 from a blank map of Asia . To test the null hypothesis of no association between genotype and climate , a Fisher's exact test was performed at α = 0 . 05 ( IBM SPSS Statistics v 20 ) . Post-hoc analyses were then performed to determine which cell ( s ) in the table of genotype versus climate contributed the most to the statistically significant Fisher's exact test . Adjusted standardized residuals ( z-scores ) were calculated and the Bonferroni method was used to correct for multiple comparisons . The adjusted standardized residual values were then compared against the critical z-value ( ±1 . 96 ) for α = 0 . 05 ( IBM SPSS Statistics v 20 ) . Only GI-a , GI-b , GII and GIII were considered in these analyses , as the dataset included only three sequences each for GIV and GV . The null hypothesis of no phylogeny-trait association was further evaluated at α = 0 . 05 using the association index ( AI ) , parsimony score ( PS ) , unique fraction ( UniFrac ) , nearest taxa ( NT ) , net relatedness ( NR ) , phylogenetic diversity ( PD ) and maximum exclusive single-state clade size ( MC ) statistics calculated from the posterior set of trees generated by BEAST in Befi-BaTS v 0 . 1 . 1 [62] . Nonsynonymous substitutions involved in the phylogenetic divergence of the five genotypes of JEV were identified within the E protein alignment . The E gene alignment was evaluated for statistically significant evidence of positive selection ( ratio of nonsynonymous to synonymous nucleotide substitutions [dN/dS] >1; p<0 . 05 ) using the single-likelihood ancestor counting ( SLAC ) , fixed effects likelihood ( FEL ) and internal FEL ( IFEL ) methods [63] available on the Datamonkey webserver [64] . All analyses of positive selection utilized a NJ phylogeny and the reversible nucleotide substitution model . Evidence of directional selection within the E protein alignment was evaluated using the directional evolution of protein sequences ( DEPS ) [65] method implemented in HyPhy v 2 . 0 [66] . The DEPS method utilized a Bayesian phylogeny and the Jones , Taylor , Thorton amino acid substitution model to assess for the presence of statistically significant shifts in amino acid residue frequencies ( p<0 . 05 ) and/or a statistically significant large number of substitutions toward a particular residue ( BF>100 ) .
Of the 487 isolates in the JEV dataset ( Table S2 ) , the majority belonged to GI-b and GIII , the most common viral host was the mosquito , the most frequent decade of isolation was the 2000s , the most common country of virus isolation was Japan and the majority of isolates were collected in temperate climates ( Table 1 ) . The overall median evolutionary rates estimated from the JEV E gene country and climate datasets were 5 . 33×10−4 ( 95% HPD: 3 . 92×10−4 , 6 . 52×10−4 ) and 5 . 51×10−4 ( 95% HPD: 4 . 24×10−4 , 6 . 67×10−4 ) substitutions/site/year , respectively . These estimates are slightly higher and the 95% HPD intervals are slightly wider compared to estimates previously obtained from a dataset of 35 JEV ORF sequences ( mean: 4 . 35×10−4 substitutions/site/year , 95% HPD: 3 . 49×10−4 , 5 . 30×10−4 ) [33] . This variation was likely due to the fact that more temporal signal can be extracted from longer alignments . Figures 1 and 2 show the geographical distribution of the JEV sequences included in this study according to the country and climate of collection , respectively . Country and climate Bayesian MCC phylogenies are shown in Figures 3 and 4 , respectively . As expected , the topologies of the BEAST phylogenies are supported by the NJ and ML phylogenies ( Figures S1 and S2 ) , and are similar to recently published phylogenies generated from both ORF and E gene sequence information for GI-V of the virus [31] , [32] , [33] . All four of the phylogenies inferred in this study support the division of GI into two clusters , GI-a and GI-b , where the GI-a clade consists of 15 isolates sampled in Cambodia , Thailand and Australia between 1967 and 2005 and GI-b includes 219 isolates sampled from Vietnam , Thailand , Japan , Korea , China and Taiwan between 1979 and 2009 . Estimated dates of the MRCA and state PP values in support of each of the 12 countries are presented in Table 2 for the key nodes within the country Bayesian MCC phylogeny ( Figure 3 ) , and estimated dates of the MRCA and state PP values in support of tropical and temperate climates of divergence are presented in Table 3 for the key nodes within the climate Bayesian MCC phylogeny ( Figure 4 ) . The country Bayesian MCC phylogeny ( Figure 3 ) and the country map ( Figure 1 ) show that GV includes three isolates sampled in China , South Korea and Malaysia between 1952 and 2010 , GIV includes three isolates sampled in Indonesia between 1980 and 1981 , GIII includes 234 isolates sampled in China , India , Indonesia , Japan , Korea , Sri Lanka , Taiwan and Vietnam between 1935 and 2009 , GII includes 28 isolates sampled in Australia , Indonesia , Korea and Malaysia between 1951 and 1999 , GI-a includes 15 isolates sampled in Cambodia , Thailand and Australia between 1967 and 2005 , and GI-b includes 219 isolates sampled from Vietnam , Thailand , Japan , Korea , China and Taiwan between 1979 and 2009 . Phylogeographic analysis estimated that the date of the MRCA of JEV lies between 1506 and 1704 with a posterior probability of 50% , and between 1022 and 1800 with a posterior probability of 95% ( median: 1553 , 50% HPD: 1506 , 1704; 95% HPD: 1022 , 1800 ) . These estimates are consistent with those recently inferred using a dataset of 35 JEV ORF sequences ( mean: 1559; 95% HPD: 1509 , 1635 [33] . The increased width of the 95% HPD intervals for the date of the MRCA of JEV for the E gene sequence dataset compared to the ORF gene sequence dataset can again be attributed to the fact that more temporal signal can be extracted from longer alignments . In agreement with previous suggestions regarding the origin of JEV [34] , the root ( MRCA of JEV ) state PP values for all locations range between 0 . 03 and 0 . 19 , with the highest state posterior probability values corresponding to Malaysia and Indonesia ( 0 . 19 and 0 . 18 , respectively ) . Of the five genotypes of JEV , the MRCA of GIII occurred earliest in time ( median: 1894; 95% HPD: 1857 , 1916 ) possibly in Japan ( state PP: 0 . 57 ) , followed by the MRCA of GV ( median: 1902; 95% HPD: 1813 , 1938 ) possibly in Malaysia ( state PP: 0 . 40 ) , the MRCA of GII ( median: 1913; 95% HPD: 1867 , 1939 ) possibly in Indonesia ( state PP: 0 . 40 ) , the MRCA of GI ( median: 1936; 95% HPD: 1908 , 1957 ) possibly in Vietnam ( state PP: 0 . 44 ) and , most recently , the MRCA of GIV ( median: 1971; 95% HPD: 1948 , 1977 ) possibly in Indonesia ( state PP: 0 . 98 ) ( Figure 3 , Table 2 ) . Within GI , the MRCA of GI-a occurred first ( median: 1949; 95% HPD: 1927 , 1962 ) possibly in Thailand ( state PP: 0 . 43 ) , followed by the MRCA of GI-b ( median: 1961; 95% HPD: 1941 , 1971 ) possibly in Vietnam ( state PP: 0 . 56 ) ( Figure 3 , Table 2 ) . The MRCA of the recently emerged GV isolates ( XZ0934 [China , 2009] and 10-1827 [South Korea , 2010]; node PP: 1 . 00 ) occurred recently ( median: 1997; 95% HPD: 1982 , 2006 ) possibly in Korea ( state PP: 0 . 51 ) ( Figure 3 , Table 2 ) . The most striking observation from the climate Bayesian MCC phylogeny ( Figure 4 ) and the climate map ( Figure 2 ) is that GV includes isolates sampled from temperate and tropical locations , GIV includes isolates sampled from only tropical locations , GIII and GI-b include isolates sampled primarily from temperate locations , and GII and GI-a include isolates sampled primarily from tropical locations . The posterior probabilities for a tropical or temperate climate at the root of the tree were approximately equal ( tropical state PP: 0 . 53 , temperate state PP: 0 . 47 ) ( Figure 4 , Table 3 ) . The MRCA of GIII ( state PP: 0 . 97 ) and the recently emerged GV isolates ( state PP: 0 . 99 ) was most likely in temperate Asia , while the MRCA of GV ( state PP: 0 . 65 ) , GII ( state PP: 0 . 77 ) , GI ( state PP: 0 . 87 ) , GI-a ( state PP: 0 . 97 ) , GI-b ( state PP: 0 . 67 ) and GIV ( state PP: 0 . 99 ) was most likely in tropical Asia ( Figure 4 , Table 3 ) . A Fisher's exact test was used to statistically evaluate the observed relationship between genotype and climate . Based on α = 0 . 05 , we rejected the null hypothesis of no genotype-climate association and concluded that there was a statistically significant relationship between genotype and climate ( Fisher's exact test: 173 . 48; exact two-sided p-value: 0 . 000 ) . Post-hoc analysis revealed that GIII included significantly more isolates sampled from temperate climates than expected ( adjusted standardized residual: 4 . 0 ) , GII included significantly more isolates sampled from tropical climates than expected ( adjusted standardized residual: 12 . 4 ) , GI-a included significantly more isolates sampled from tropical climates than expected ( adjusted standardized residual: 9 . 3 ) and GI-b included significantly more isolates sampled from temperate climates than expected ( adjusted standardized residual: 5 . 2 ) . The phylogeny-trait association test of genotype-climate phylogenetic structure also failed to reject the null hypothesis of no association between genotype and climate ( Table S3 ) ; thereby , providing further evidence that JEV genotype is associated with climate . Forty-four genotype-defining nonsynonymous substitutions were identified within the E protein ( Table S4 ) , 36 of which were GV-specific ( 13 non-conservative ) . Selection analyses were then performed to determine if any of these genotype-defining nonsynonymous substitutions might have played a role in the adaptation of the viral genotypes to their respective environments . No positively selected sites were identified using the SLAC , FEL and IFEL methods . However , the DEPS method revealed elevated substitution rates towards seven residues ( Table S5 ) and thirteen sites were identified to be involved in this directional evolution ( Table S6 ) . One of these directionally selected sites ( 129; preferred residue: M ) corresponded to the site of a genotype-defining nonsynonymous substitution ( 129 [GV: I , GIV: T , GIII: T , GII: T , GI: M] ) .
In accordance with previous analyses , we estimated that JEV originated in the Indonesia-Malaysia region [34] around the 16th century [33] . However , as expected for a node so far back in time , the posterior probability values in support of the origination of JEV in the Indonesia-Malaysia region were low . Nevertheless , as emphasized previously , all virus genotypes have been found in the Indonesia-Malaysia region and large epidemics suggestive of JEV have never been reported to occur in this region [34] . These lines of evidence are consistent with the virus having evolved in the Indonesia-Malaysia region [34] . Interestingly , based on the results of an amino acid signature analysis , others have suggested that Asian JEV and Australian Murray Valley encephalitis virus may have evolved from a virus related to the African Usutu virus in the Southeast Asia-Australasia region [67] . Phylogeographic analysis estimated that GIII evolved in temperate Asia ( Japan ) around the late 19th century . These estimates are coincident with the first reported summer epidemics of encephalitis suggestive of JE , which occurred in 1871 in Japan [1] . Following the isolation of the prototype Nakayama strain of JEV ( a GIII virus ) from Japan in 1935 , GIII has been found throughout most of Asia including China , India , Indonesia , Korea , Japan , Sri Lanka , Taiwan and Vietnam . Statistical analyses indicated that GIII did indeed include significantly more temperate isolates than expected under the null hypothesis of no association between genotype and climate . The paucity of GIII viruses sampled from tropical regions and the genetic relatedness of GIII viruses sampled years apart suggests that the annual re-introduction of GIII viruses from tropical regions to temperate regions by migratory birds or wind-blow mosquitoes does not seem to play a large role in the epidemiology of GIII . Rather , GIII is most likely maintained year-to-year by hibernating mosquitoes , vertical transmission in mosquitoes , poikilothermic vertebrates and/or bats . In support of this hypothesis , JEV has been isolated from overwintering Culex sp . in temperate Asia [68] , virus transmission by Culex spp . was shown following experimental hibernation [69] , and vertical transmission of JEV in Culex spp . and Armigeres sp . has been experimentally demonstrated [70] , [71] . Furthermore , antibody to JEV has been detected in several poikilothermic vertebrates [71] , [72] , [73] , [74] , experimentally JEV-infected lizards were able to maintain the virus throughout the winter [72] , and experimental transmission of JEV has been shown from infected mosquitoes to uninfected lizards and from infected lizards to mice through mosquitoes [72] . In temperate Asia , JEV has been isolated from several bat species [73] , [74] and JEV-infected bats subjected to experimental hibernation were able to maintain their viremias for over 100 days [75] . The MRCA of the three GV sequences was estimated to have existed in the early 20th century . Although the node states were largely influenced by the small number of GV sequences ( n = 3 ) , the mass of posterior probability supporting the location of GV evolution corresponds to tropical Asia , specifically Malaysia . In Malaysia , JEV was first described in the 1940s when an outbreak occurred during the Second World War among British prisoners of war [76] . It is possible that GV may have circulated undetected in tropical Asia for much longer , causing only sporadic cases of encephalitis that were mistaken for cerebral malaria or other encephalitic diseases . Surprisingly , after almost 60 years of undetected virus circulation , a pool of Cx . tritaeniorhynchus collected in the Tibetan Province of China in 2009 yielded the GV XZ0934 isolate [31] and a pool of Culex bitaeniorhynchus collected in South Korea in 2010 yielded the GV 10-1827 isolate [32] . The MRCA of the XZ0934 and 10-1827 isolates was estimated to have occurred sometime within the last 27 years in temperate Asia . Interestingly , despite surveillance neither JEV nor Cx . tritaeniorhynchus had been detected in Tibet prior to 2009 [77] , suggesting that GV of JEV may have entered Tibet shortly before it was initially isolated in 2009 . It is possible that GV arrived in Tibet via JEV-infected migratory birds or perhaps by wind-blown mosquitoes . The three GV viruses shared 36 genotype-defining nonsynonymous substitutions within the E protein , 13 of which were non-conservative . This is consistent with the Muar strain's distinct serological classification based on its reactivity with a set of monoclonal antibodies [78] . None of the GV isolates have been characterized using polyclonal antibodies derived from other members of the JEV serocomplex . Such studies may provide interesting information regarding the antigenic relationships between this ancestral JEV genotype and other closely related viruses , such as Murray Valley encephalitis virus and Usutu virus . We estimated that GII evolved in tropical Asia around the early 20th century . The Bennett isolate , made in Korea circa 1951 , represents the only example of a GII virus collected outside of tropical Asia [40] . As extensive surveillance for JEV has been performed in temperate Asia , this single genotype isolation event likely coincided with a GII epidemic focus that quickly died off [40] . Therefore , as predicted , statistical analyses demonstrated that GII included significantly more isolates sampled from temperate regions than expected . It was estimated that GI , GI-a and GI-b all emerged in tropical Asia around the mid 20th century . Statistical analyses demonstrated that GI-a included significantly more isolates sampled from tropical regions , and GI-b included significantly more isolates sampled from temperate regions , than expected under the null hypothesis of no association between climate and genotype . Like GIII , GI-b may be maintained in temperate Asia throughout the winter months in hibernating mosquitoes , vertical transmission in mosquitoes , poikilothermic vertebrates , and/or bats . This suggests that the spread and establishment of GI-b throughout Asia may have been due to its ability to efficiently overwinter in temperate Asia . The phylogenetic divergence of GI is defined by a non-conservative threonine to methionine substitution at site 129 of the E protein . This GI-defining substitution was also found to be under directional selection . Although substitutions at this site within the E protein of JEV have not yet been associated with phenotypic alterations , this substitution alone or in combination with substitutions in other regions of the genome may have provided a phenotypic advantage to GI viruses that led to the spread and establishment of this genotype throughout Asia . The MRCA of the three GIV sequences was estimated to have existed in the late 20th century . GIV includes seven isolates ( only three of these include E gene sequence information ) collected from mosquitoes only on three islands encompassing the Indonesian archipelago between 1980 and 1981 . The reasons why GIV appears to be confined to Indonesia are unknown , but could be due to a number of reasons . For example , there could be a narrow host/vector range for GIV , the vector competence of Cx . tritaeniorhynchus for GIV may be low , the primary vector of GIV may be a mosquito that is confined to Indonesia , the replicative ability of GIV in birds may be low , and/or the GIV transmission cycle may involve a non-migratory amplifying host [42] . Although valuable information was obtained from our phylogeographic study , the data should be interpreted in light of its limitations . Isolations of JEV prior to the 1970s were primarily made from humans residing in China and Japan in response to epidemic transmission of the virus . After 1970 , the majority of JEV isolates were made from mosquitoes and swine throughout many prefectures of Japan ( due to rare cases of JEV following introduction of an effective vaccination program ) and provinces of China as part of yearly routine surveillance for JEV . Therefore , it is unlikely that the dataset is largely biased by sequences sampled from local endemic foci that would have confounded the reported genotype-climate association . Due to the small number of ORF sequence data for JEV , we utilized a dataset consisting of sequence information derived from the E gene . While the E gene of JEV was found to be a good evolutionary proxy of the ORF , we were not able to assess whether diversifying and/or directional selection in other regions of the genome may have played a role in the adaptation of the viral genotypes to their respective environments . Finally , phenotypic characterization of the five genotypes of JEV has yet to be performed and would provide further evidence to support the genotype-climate association . By applying Bayesian phylogeographic , categorical data analysis and phylogeny-trait association techniques to a large JEV sequence dataset we have demonstrated that GIII and GI-b are temperate genotypes maintained year-round in northern latitudes likely by either hibernating mosquitoes , vertical transmission in mosquitoes , poikilothermic vertebrates and/or bats . In contrast , GI-a and GII are tropical genotypes likely maintained via mosquito-avian and/or mosquito-swine transmission cycles . This suggests that the spread and establishment of GI-b throughout Asia may have been due to its ability to efficiently overwinter in temperate Asia . As highlighted by the recent emergence of West Nile virus into the western hemisphere [79] and Usutu virus into the European continent [80] , the invasion of JEV into previously unoccupied regions is a real threat . Many areas of the world have JEV-competent vectors and waterbirds , and unlike West Nile and Usutu viruses , JEV also utilizes domestic swine as amplifying hosts , which can drive epidemics by producing an abundance of infected mosquitoes . | Although Japanese encephalitis virus ( JEV ) is a major cause of death and disability throughout tropical and temperate Asia , little is known about the evolution , geographical distribution and epidemiology of the five JEV genotypes ( genetically distinct groups ) . To address this gap in our knowledge , we performed a genetic-based geographical analysis using the largest JEV sequence dataset assembled to date , including 487 viral sequences sampled from 12 countries over 75 years . We showed that both the newly and previously dominant genotypes of JEV are associated with temperate climates and are maintained throughout the cold winter months in northern Asia , likely by hibernating mosquitoes ( survive throughout the winter ) , vertical transmission in mosquitoes ( female to offspring ) , cold-blooded vertebrates and/or bats . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"evolutionary",
"ecology",
"ecology",
"neglected",
"tropical",
"diseases",
"biology",
"japanese",
"encephalitis"
] | 2013 | Phylogeography of Japanese Encephalitis Virus: Genotype Is Associated with Climate |
Neurocysticercosis , infection of the brain with larvae of Taenia solium ( pork tapeworm ) , is one of several forms of human cysticercosis caused by this organism . We investigated the role of albendazole and praziquantel in the treatment of patients with parenchymal neurocysticercosis by performing a meta-analysis of comparative trials of their effectiveness and safety . We performed a search in the PubMed database , Cochrane Database of Controlled Trials , and in references of relevant articles . Six studies were included in the meta-analysis . Albendazole was associated with better control of seizures than praziquantel in the pooled data analysis , when the generic inverse variance method was used to combine the incidence of seizure control in the included trials ( patients without seizures/[patients×years at risk] ) ( 156 patients in 4 studies , point effect estimate [incidence rate ratio] = 4 . 94 , 95% confidence interval 2 . 45–9 . 98 ) . In addition , albendazole was associated with better effectiveness than praziquantel in the total disappearance of cysts ( 335 patients in 6 studies , random effects model , OR = 2 . 30 , 95% CI 1 . 06–5 . 00 ) . There was no difference between albendazole and praziquantel in reduction of cysts , proportion of patients with adverse events , and development of intracranial hypertension due to the administered therapy . A critical review of the available data from comparative trials suggests that albendazole is more effective than praziquantel regarding clinically important outcomes in patients with neurocysticercosis . Nevertheless , given the relative scarcity of trials , more comparative interventional studies—especially randomized controlled trials—are required to draw a safe conclusion about the best regimen for the treatment of patients with parenchymal neurocysticercosis .
Neurocysticercosis is a parasitic disease caused by the larval form of Taenia solium , known as pork tapeworm , when the larvae lodge in the central nervous system ( CNS ) . It happens when human ingests the eggs , acting as the intermediate host in the life cycle of T . solium . The eggs hatch in the intestine and the embrya penetrate the intestinal wall and are distributed via the blood , anchoring in the CNS as a larval form of the parasite [1] . With T . solium parasitosis , both self-reinfection and infection of household members are common . Neurocysticercosis is mosst commonly found among members of agricultural societies with poor sanitary conditions and economies based on breeding livestock , especially pigs , with low hygiene standards [2] . However , it has also started to emerge in developed countries , as a result of immigration from endemic to nonendemic areas [3] . Its natural pool lies mainly in Latin America , sub-Saharan Africa , and Southeast Asia , and is an important cause of morbidity among local populations [2] . Neurocysticercosis is divided into four categories depending on the anatomical locus in which the larvae lodge—cerebral or parenchymal , subarachnoid or cisternal , intraventricular , and spinal [1] . The most common clinical sign of neurocysticercosis is epilepsy of any type , which is usually late-onset; this sign is typically found in parenchymal neurocysticercosis . Other common signs are focal neurological deficits , cerebellar or brainstem signs , signs of increased intracranial pressure , meningoencephalitic signs , dementia , or even death [4] . The standard therapeutic intervention was surgery until the development of cysticidal agents , the most common being praziquantel and albendazole [5] . Although there have been many clinical trials testing these drugs , controversy remains about their therapeutic value [5] . The reasons for this dispute include the severity of adverse effects , the actual reduction of cysts , and the subsequent control of seizures . This disagreement seems to have been resolved after the recent publication of a meta-analysis that shows the superiority of these agents compared to placebo [6] . We sought to investigate which of the two agents are preferable in the treatment of neurocysticercosis . Some studies have been published on this issue , although they mostly examine small numbers of patients . Specifically , we investigated the role of albendazole versus praziquantel in the treatment of patients with parenchymal neurocysticercosis by performing a meta-analysis of comparative trials [7] of their effectiveness and safety .
The studies for our meta-analysis were obtained from the PubMed database , Cochrane Database of Controlled Trials , and from references of relevant articles . Search terms included “albendazole” , “praziquantel” , “neurocysticercosis” , and “Taenia solium” . Although the search was performed without limitation on the language of publications , the evaluable studies were published in English , French , German , and Italian . There was no limitation on the year of publication . Two independent reviewers ( DKM and GP ) performed the search and selected the studies that were relevant to the scope of our meta-analysis . Any discrepancy or disagreement between the reviewers was resolved by consensus in meetings involving all authors . A study was considered eligible if ( 1 ) it was a prospective trial , ( 2 ) it compared albendazole with praziquantel for the treatment of patients with neurocysticercosis , ( 3 ) it examined the partial or total disappearance of cysts and/or control of seizures , and ( 4 ) if it included patients infected with parasites in their cystic stage without perilesional inflammation . Studies using concomitant drugs such as corticosteroids , analgesics , and anticonvulsive drugs were not excluded . The following data were extracted from each study: year of publication , study design , population of the study , therapeutic regimens used , concomitant drugs , number of patients , follow-up period , patients having control of seizures , proportion of cyst reduction , disappearance of cysts , total toxicity , and patients presenting intracranial hypertension as a side effect . A quality review of each randomized controlled trial ( RCT ) included in our analysis was performed by using the Jadad score , which examines whether there is randomization , blinding , and information on withdrawals in the study , and evaluates the appropriateness of randomization and blinding , if present . One point was awarded for the presence of each of the first 3 criteria , whereas the last 2 criteria could take the values of −1 ( inappropriate ) , 0 ( no data ) , and +1 ( appropriate ) [8] , [9] . Thus , the maximum score for a study was 5 , and a score more than 2 points denoted an adequate RCT according to the methodology . The reviewers calculated the score of each RCT independently . Any disagreement was resolved after consensus among all authors . The primary outcome was the proportion of patients with controlled seizures . Secondary outcomes were the reduction of cysts in all of the examined patients , the proportion of patients with total disappearance of cysts , the proportion of patients with adverse events related with the administered antihelminthic drugs , and the proportion of patients with intracranial hypertension as a side effect caused by the administered drugs . A patient was considered as having total control of seizures when there had been no seizures during the follow-up period . A patient was considered as having total disappearance of cysts when this outcome had been achieved after only one course of administered chemotherapy and without any surgical intervention at the follow-up CT scan , performed in a time frame of 3 to 6 months after the end of therapy . The reduction of cysts was defined as the proportion of the number of cysts that had resolved by the follow-up evaluation ( numerator ) , which varied from 3 to 6 months post-therapy , divided by the number of cysts at baseline ( denominator ) . Adverse events included any type of adverse event reported in the included studies . Statistical analyses were performed using the “Review Manager 4 . 2” software and the SPSS 15 . 0 statistical software . The heterogeneity between studies was assessed by using the I2 test and χ2 test; for the χ2 test , p<0 . 10 was considered statistically significant in the analysis of heterogeneity [10] . Small-study bias was assessed by the funnel plot method [11] . Pooled odds ratios ( ORs ) and 95% confidence intervals ( CIs ) for all primary and secondary outcomes were calculated by using both the Mantel-Haenszel [12] fixed effect model and the DerSimonian-Laird random effects model [13] . For all analyses , results from the fixed effect model are presented only when there was no heterogeneity between studies; otherwise , results from the random effects model are presented . For the analyses of proportions of the reduction of cysts , we used a linear regression model in which the percentage of reduction of cysts for each treatment arm in the included studies was the dependent variable , and the administered drug was the independent variable . With this model , a beta ( β ) coefficient of the independent variable was calculated as well as the 95% confidence interval ( CI ) of the coefficient . For the analyses of seizure control for which the follow up period varied , we combined the logarithms of the rate ratios across the included trials ( patients with outcome/[patients×years at risk] ) using the generic inverse variance method .
Figure 1 is a flow diagram describing the process of study selection . We identified 103 potentially evaluable papers , 91 of which were excluded because they were reviews , case reports , letters or editorials , laboratory studies , small series of patients , retrospective studies , and meta-analyses that examined a different aspect of neurocysticercosis than the comparison between praziquantel and albendazole . Of the remaining 12 potentially evaluable papers , 2 studies were excluded because they included patients with neurocysticercosis that was not parenchymal , 1 because the majority of the enrolled patients had mixed living and calcified cysts , 1 because the enrolled patients were all put in the same group without providing separate data for each antiparasitic agent , and 2 because they were subsets of other larger trials . Thus , 6 trials were included in our meta-analysis [14]–[19] . The assessment of the evaluable studies according to the Jadad score was performed only for the 2 out of 6 studies [15] , [19] . The rest of the studies were prospective [14] , [16]–[18] but not RCTs . Thus , quality assessment of these trials using Jadad could not be done . The studies differed in the administered dosing and duration of therapy for albendazole and praziquantel ( Table 1 ) . Most of the researchers administered 15 mg/kg/d of body weight of albendazole , but the duration of therapy varied from 8 days to a month [14]–[16] , [18] , [19] . In only one study albendazole was administered at a dosage of 20 mg/kg/d for 21 days [17] . We pooled these data , as the administration of albendazole for 7 days is as effective as for longer periods of therapy [20] . There was notable variation in the duration of praziquantel therapy , extending from a single day to 3 weeks . In all of the studies the dosage of praziquantel was 50 mg/kg/d , except one study in which praziquantel was administered at a dosage of 100 mg/kg in 3 divided doses at 2-hour intervals for a single day [14] . We pooled these data , as the administration of praziquantel for a single day is as effective as for longer periods of therapy [21]–[23] . Data on the complete control of seizures in patients with neurocysticercosis treated with albendazole or praziquantel were reported in 4 out of 6 studies ( Table 2 ) [14]–[17] . One study reported a statistically significant effects in favor of albendazole , as reported in the crude data provided in the study [17] . To overcome the variation in the follow-up periods , we used the generic inverse variance method to combine the incidence of seizure control ( patients without seizures/[patients×years at risk] ) of the included trials ( Table 2 ) . Albendazole was associated with better control of seizures in comparison with praziquantel in the pooled data analysis ( 156 patients , random effects model [I2 = 51 . 2%] , point effect estimate [incidence rate ratio] = 4 . 94 [seizure-free persons/person-years] , 95% CI 2 . 45–9 . 98 , Figure 2 ) . Data on the reduction of the total number of cysts from baseline to follow-up are reported in 5 out of 6 studies ( Table 2 ) [14]–[17] , [19] . A linear regression model of the proportion of reduction of cysts and the administration of albendazole or praziquantel yielded a beta coefficient ( β ) = 0 . 22 ( standard error [SE] = 0 . 113 ) with 95% CI −0 . 05 to 0 . 48 . The analysis included a total of 301 patients with 2565 cysts . Hence , there was no statistically significant difference in the proportion of the reduction of cysts between albendazole and praziquantel for the treatment of neurocysticercosis . In addition , in a sensitivity analysis excluding the data reported in the RCT by Sotelo et al [19] which comprised almost one half of the total number of cysts , there was no statistically significant difference in the proportion of the reduction of cysts between albendazole and praziquantel for the treatment of neurocysticercosis ( β = 0 . 15 [SE = 0 . 18] , 95% CI −0 . 30 to 0 . 59 ) . The analysis included a total of 187 patients with 1342 cysts . Data on the total disappearance of cysts are reported in all 6 studies ( Table 2 ) [14]–[19] . Albendazole was associated with greater efficacy than praziquantel in the total disappearance of cysts ( 335 patients , random effects model ( χ2-test p = 0 . 07 , I2 = 50 . 3% ) , OR = 2 . 30 , 95% CI 1 . 06–5 . 00 , Figure 3 ) . Since in the study by Cruz et al [18] it is not clear whether the patients with cystic lesions also had lesions involving other stages of the infection , we performed a sensitivity analysis without the aforementioned study , in which albendazole was more effective than praziquantel in inducing the total disappearance of cysts ( 301 patients , random effects model ( χ2-test p = 0 . 05 , I2 = 58 . 1% ) , OR = 2 . 62 , 95% CI 1 . 09–6 . 32 ) . We also performed a sensitivity analysis excluding data reported in the RCT by Sotelo et al [19] , which included almost one-third of the total number of patients in this meta-analysis and showed statistical significance . There was no difference between the two regimens in inducing the total disappearance of cysts ( 221 patients , random effects model ( χ2-test p = 0 . 08 , I2 = 52 . 5% ) , OR = 2 . 20 , 95% CI 0 . 79–6 . 13 ) . Data about mortality are reported in all 6 studies ( Table 2 ) [14]–[19] . One death was reported in by Takayanagui et al [17] due to increased intracranial pressure . These data were not adequate to allow a meaningful analysis . Data about patients with adverse events are reported in 5 out of 6 studies ( Table 2 ) [14] , [15] , [17]–[19] . Albendazole and praziquantel did not differ in the proportion of patients with adverse events ( 388 patients , random effects model [χ2-test p = 0 . 06 , I2 = 59 . 9%] , OR = 0 . 67 , 95% CI 0 . 26–1 . 69 ) . Data on intracranial hypertension developing as a consequence of the regimens administered are reported in 4 studies ( Table 2 ) [14] , [16]–[18] . There was no difference in the development of intracranial hypertension due to the administered therapy between albendazole and praziquantel ( 179 patients , fixed effect model [χ2-test p value = 0 . 58 , I2 = 0%] , OR = 0 . 31 , 95% CI 0 . 05–2 . 09 ) .
Neurocysticercosis is an endemic disease in many developing countries , and it may expand to the developed world , mainly as a result of immigration . Estimations report around 50 million new cases worldwide [24] . To our knowledge , until now the guidelines for the treatment of cysticercosis are the result of a consensus by a panel of experts in the subject [25] . Specifically , for viable parenchymal cysts the recommendations are based on evidence obtained from multiple case series with or without intervention , including dramatic results in uncontrolled experiments ( level II-3 recommendation , which is considered a weak category of evidence ) , and on opinions of respected authorities , based on clinical experience , descriptive studies , and case reports or reports of expert committees ( level III recommendation ) . Although these recommendations support the use of antiparasitic treatment , they do not point to either albendazole nor praziquantel as the drug of choice for this type of neurocysticercosis . In a recent meta-analysis performed by Del Brutto et al . [6] it was suggested that , compared to placebo , cysticidal drug therapy results in better resolution of colloidal and vesicular cysticerci , lower risk for recurrence of seizures in patients with colloidal cysticerci , and a reduction in the rate of generalized seizures in patients with vesicular cysticerci . However , there has not yet been a meta-analysis comparing the effectiveness and safety of albendazole and praziquantel in patients with neurocysticercosis . The outcomes in our meta-analysis suggest that albendazole is more effective than praziquantel in controlling seizures in the affected patients and in leading to the total disappearance of cysts and , subsequently , the cure of patients with neurocysticercosis . However , in the sensitivity analysis of the total disappearance of cysts , excluding the study by Sotelo et al [19] , no significant difference was found between the drugs , although the odds ratio was rather similar to the analysis that included the study by Sotelo et al . [19] . This loss of statistical significance can be explained by the loss of power in the sensitivity analysis due to exclusion of the aforementioned study . Regarding other outcomes , there have been no statistically significant differences between albendazole and praziquantel in reduction of total number of cysts , mortality , total adverse events , and development of intracranial hypertension due to the administered therapeutic agents . Control of seizures and total disappearance of cysts were chosen as outcomes in our meta-analysis , because they are easily defined and quantitatively measured . In addition , new-onset seizures are among the most common symptoms that lead patients to seek medical care , and their resolution is one of the major goals of therapy . In the analyses of outcomes we did not perform sensitivity analyses that excluded the study by Medina et al [16] , in which patients did not receive corticosteroids . Since it is the only study with this characteristic , one may suggest that it could cause bias . It might be speculated that the absence of corticosteroids could interfere with the kinetics of the administered antihelminthics , and cause an increase in the rate of the adverse events . However , all the outcomes included in this study did not differ from the results of the other trials; adverse events are not reported in this study . The reduced effectiveness of praziquantel could be explained by the interaction between praziquantel and corticosteroids , which results in decreased serum concentration of praziquantel [26] . Also , praziquantel interacts with anti-epileptic drugs [27] , [28] , thus altering its bioavailability . In contrast , corticosteroids interact with albendazole by decreasing the rate of elimination of albendazole sulfoxide , which is the active metabolite of albendazole , thus increasing serum concentrations of albendazole sulfoxide [29] , [30] . Often , the first few days after the administration of antiparasitic agents to patients with neurocysticercosis there is a recrudescence of neurological symptoms , most importantly decompensation of intracranial pressure and the onset of seizures or worsening of pre-existing ones , owing to peri-lesional inflammation due to degeneration of the parasite; this condition can be life-threatening . The severity of inflammation is proportional to the parasitic burden , resulting in more severe manifestations in individuals with greater cyst loads [31] . A common approach to ameliorating this problem is the concomitant administration of corticosteroids to reduce edema , the inflammatory response , and intracranial hypertension [32] . Special attention should be paid to patients with high cyst loads , to whom the administered antiparasitic treatment causes an abrupt degeneration of cysts that may lead to severe inflammation and seizures [5] . In such cases corticosteroids should be administered before the antiparasitic agents . The single death reported in the study by Takayanagui et al [17] ( the only death among patients of all trials included in this meta-analysis ) was the result of increased intracranial pressure , which , however , pre-existed at the beginning of the trial . In 5 out of 6 studies included in our meta-analysis , corticosteroids were administered to patients [14] , [15] , [17]–[19] . Only in the study by Medina et al [16] were corticosteroids not administered; adverse events were not reported in this study . It is believed by several experts that many cysts degenerate spontaneously over time , which may lead to the conclusion that the results of the evaluable studies may be biased [33] . Since it is not clear up to what extent this opinion is true , we analyzed studies that included patients with cystic lesions without perilesional enhancements or other evidence of surrounding inflammation , as evidence of a possible degenerative process , to rule out such a possibility . Antihelminthic drugs are effective against viable cysts , but not on remnants , granulomas , and calcifications of dead cysts . Thus , both outcomes we chose to study—the total disappearance of cysts and reduction of cysts—are useful indicators of the effectiveness of the administered therapy , because they estimate the effectiveness of the administered agents for lesions on which the agents are active . There are some limitations in our meta-analysis that should be considered . First , one may claim that the number of the studies and the number of patients are too small to allow a definitive conclusion regarding the results of the compared therapies . This small sample size is important because it leads to large confidence intervals . In addition , publication bias cannot be appropriately assessed in a small set of studies . Also , among the studies selected there are only 2 RCTs [15] , [19] in a total of 6 comparative trials , which prevents us from applying the usually applied methodology in obtaining an overall quality assessment of the included studies [8] . Second , there are discrepancies in the administered dosage and duration of therapy with the 2 antiparasitic agents used . Although there have been several studies aiming to establish an optimal dosage and duration of therapy , these important therapeutic parameters have not been standardized yet . We pooled all of the available data , since the dosage and the duration of therapy used in the trials included in this meta-analysis are generally accepted alternatives by the medical community . Furthermore , there were differences in the length of follow-up for the control of seizures between the studies that varied from 6 to 24 months . This fact may give rise to methodological issues regarding the validity of combining these studies without considering the duration of follow-up . Thus , we performed an analysis using the generic inverse variance method combining the incidence of seizure control in the included trials ( patients without seizures/[patients×years at risk] ) , in which the effect of different follow-up time is included . However , it should be noted that the caveat in this methodology is the assumption that the risk for seizures is constant , which is not proven . Despite the aforementioned limitations , the contribution of the meta-analysis in the literature sheds light in the subject given the scarcity of data . In summary , neurocysticercosis is a disease with a long history in humans and with many different stages . We concentrated on parenchymal neurocysticercosis with viable cysts . The recommendations suggest the administration of antiparasitic treatment with concomitant use of steroids . This meta-analysis sought to provide more accurate estimates of the comparative effectiveness and safety of albendazole and praziquantel for this common parasitic infection . Nevertheless , more studies , especially randomized controlled trials , with homogeneous regimens and long follow-up periods , are required to draw a clear conclusion about the best regimen for the treatment of patients with parenchymal neurocysticercosis . | Neurocysticercosis is a parasitic disease caused by the pork tapeworm , Taenia solium , when the larval form of the parasite lodges in the central nervous system . This disease is most commonly found among members of agricultural societies with poor sanitary conditions and economies based on breeding livestock ( especially pigs ) with low hygiene standards . It is a disease with long history in humans , and the usual therapeutic intervention was surgery until the development of antiparasitic cysticidal agents , the most common being praziquantel and albendazole . T . solium infection can take many different forms in humans , but we concentrated on parenchymal neurocysticercosis with viable cysts . A consensus statement by a panel of experts on the subject supports the use of antiparasitic treatment , but does not indicate either albendazole or praziquantel as the drug of choice for this type of neurocysticercosis , because data from single relevant clinical trials are not conclusive . We conducted a meta-analysis to further evaluate the comparative effectiveness and safety of albendazole and praziquantel for this particular type of neurocysticercosis . The outcomes of our meta-analysis suggest that albendazole is more effective than praziquantel in controlling seizures in affected patients and in leading to the total disappearance of cysts and subsequently cure of patients with neurocysticercosis . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"neurological",
"disorders/infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"infectious",
"diseases/helminth",
"infections"... | 2008 | Albendazole versus Praziquantel in the Treatment of Neurocysticercosis: A Meta-analysis of Comparative Trials |
Understanding the host immune response during cryptococcal meningitis ( CM ) is of critical importance for the development of immunomodulatory therapies . We profiled the cerebrospinal fluid ( CSF ) immune-response in ninety patients with HIV-associated CM , and examined associations between immune phenotype and clinical outcome . CSF cytokine , chemokine , and macrophage activation marker concentrations were assayed at disease presentation , and associations between these parameters and microbiological and clinical outcomes were examined using principal component analysis ( PCA ) . PCA demonstrated a co-correlated CSF cytokine and chemokine response consisting primarily of Th1 , Th2 , and Th17-type cytokines . The presence of this CSF cytokine response was associated with evidence of increased macrophage activation , more rapid clearance of Cryptococci from CSF , and survival at 2 weeks . The key components of this protective immune-response were interleukin ( IL ) -6 and interferon-γ , IL-4 , IL-10 and IL-17 levels also made a modest positive contribution to the PC1 score . A second component of co-correlated chemokines was identified by PCA , consisting primarily of monocyte chemotactic protein-1 ( MCP-1 ) and macrophage inflammatory protein-1α ( MIP-1α ) . High CSF chemokine concentrations were associated with low peripheral CD4 cell counts and CSF lymphocyte counts and were predictive of immune reconstitution inflammatory syndrome ( IRIS ) . In conclusion CSF cytokine and chemokine profiles predict risk of early mortality and IRIS in HIV-associated CM . We speculate that the presence of even minimal Cryptococcus-specific Th1-type CD4+ T-cell responses lead to increased recruitment of circulating lymphocytes and monocytes into the central nervous system ( CNS ) , more effective activation of CNS macrophages and microglial cells , and faster organism clearance; while high CNS chemokine levels may predispose to over recruitment or inappropriate recruitment of immune cells to the CNS and IRIS following peripheral immune reconstitution with ART . These results provide a rational basis for future studies of immune modulation in CM , and demonstrate the potential of baseline immune profiling to identify CM patients most at risk of mortality and subsequent IRIS .
Cryptococcal meningitis ( CM ) is a leading cause of mortality in HIV-infected patients in low-resource settings [1 , 2] . Current treatments for CM are inadequate , with high associated mortality [3–5] , and high incidence of immune reconstitution inflammatory syndrome ( IRIS ) amongst those who survive and start antiretroviral therapy ( ART ) [6–8] . Understanding the host immune response to cryptococcal infection is important to enable rational development of immunomodulatory therapies , and also allow identification of patients most at risk of mortality and IRIS [3 , 7 , 9] . Much of our understanding of the immune response to Cryptococcus is derived from in-vitro and animal experiments . In these model systems effective immunity is dependent on robust CD4+ T-cell immune responses , the production of Th1-type cytokines such as interferon-γ ( IFNγ ) , and “classical” activation of effector cells such as macrophages [10–17] , leading to killing and clearance of infection . Th17-type CD4+ T-cell responses and cytokines appear to play a protective role [18–21] , whilst Th2-type responses are associated with impaired control of infection and poor outcomes [16 , 17 , 22–26] . Human data are very limited . Epidemiological evidence clearly points to the importance of adequate CD4+ T-cell mediated immunity in the control of cryptococcal infection [27 , 28] , and experimental data suggest that the phenotype of the CD4+ T-cell response to Cryptococcus neoformans influences the outcome of CM [9 , 29] . The importance of pro-inflammatory responses at the site of infection [7] , and in particular IFNγ [30] , for effective host immune responses to cryptococcal infection in HIV-infected patients has been reported . Preliminary data have not shown the Th1 / Th2 dichotomy seen in some mouse models , likely reflecting differences between carefully controlled animal models of cryptococcal infection and the complex situation in HIV-infected patients with heterogeneous immune status and organism burden [29] , and no data are available on the role of Th17-type cytokines or the magnitude or phenotype of innate effector cell activation during infection . To characterize the immune response to Cryptococcus during HIV-associated CM we measured cytokine concentrations , chemokine concentrations , and levels of macrophage activation markers in the cerebrospinal fluid ( CSF ) of ninety patients with CM enrolled in a clinical trial investigating the utility of short-course adjuvant IFNγ therapy [3] . The phenotype of the CSF immune response , and associations between the phenotype and disease burden at presentation , rate of clearance of infection , 2-week mortality , and development of IRIS were examined .
Ninety HIV-positive adults ( age ≥ 21 years ) with a first episode of cryptococcal meningitis , diagnosed by CSF India ink or cryptococcal antigen testing ( titres ≥1:1024 , Meridian Cryptococcal Latex Agglutination System; Meridian Bioscience Inc , Cincinnati , Ohio , USA ) , were enrolled sequentially between July 2007 and May 2010 into a clinical trial examining the effect of two different schedules of short-course adjuvant interferon-γ immunotherapy on the treatment of HIV-associated cryptococcal meningitis in Cape Town , South Africa [3] . The study has been described in detail elsewhere [3] . Detailed history and clinical examination findings were recorded at study enrolment . Lumbar punctures ( LPs ) with quantitative cerebrospinal fluid ( CSF ) cultures were performed on days 1 , 3 , 7 and 14 . Cryptococcal clearance ( early fungicidal activity , or EFA ) was calculated as previously described [31] . All patients had CD4+-cell counts performed at study enrollment ( FACSCount; Becton Dickinson ) . Patients were started on antiretroviral therapy consisting of stavudine/lamivudine/efavirenz at 2 to 4 weeks post commencement of antifungal therapy , and followed for one year . Mortality outcomes and the occurrence of IRIS ( diagnosed according to a standardized definition [32] ) were recorded . CSF samples collected at day 1 , 3 , 7 and 14 were centrifuged , and the supernatant frozen at -80°C for subsequent quantification of cytokine concentrations . CSF IFNγ , tumor necrosis factor-α ( TNFα ) , interleukin ( IL ) -2 , IL-4 , IL-6 , IL-8 ( chemokine ( C-X-C motif ) ligand 8 ) , IL-10 , IL-12p70 , IL-17 , IL-21 , IL-22 , IL-23 , monocyte chemotactic protein-1 ( MCP1 , or chemokine ( C-C motif ) ligand 2 ) , macrophage inflammatory protein-1α ( MIP1α , or chemokine ( C-C motif ) ligand 3 ) , RANTES ( chemokine ( C-C motif ) ligand 5 ) , granulocyte-macrophage colony-stimulating factor ( GM-CSF , or colony stimulating factor 2 ) , and vascular endothelial growth factor ( VEGF ) concentrations were measured in all patients using the Luminex multianalyte platform ( Luminex ) and Bio-Rad cytokine kits ( Bio-Rad ) [30] . The macrophage activation markers soluble CD14 ( sCD14 ) and neopterin concentrations were measured in CSF using Bio-Rad and ELISA ( ELItest Neopterin , BRAHMS Aktiengesellschaft , Hennigsdorf , Germany ) kits respectively . The enzymatic activity of arginase and protein concentration of samples were measured as previously described [33] . This analysis focuses solely on the baseline assays performed prior to administration of antifungal therapy ( results from subsequent time points are shown in S1 Fig All data are available as a supplementary S1 data file ) . Data were analysed using Stata version 13 . 0 ( StataCorp , College Station , Texas , USA ) , R version 3 . 0 . 2 ( R foundation for Statistical Computing ) , and Graphpad Prism version 6 ( Graphpad Software Inc . , San Diego , California , USA ) . Baseline CSF cytokine and chemokine concentrations were analysed using principal component analysis ( PCA ) , a method of reducing complex correlated datasets into a series of linear , non-correlated “principal components” ( PCs ) and avoiding multiple comparisons . The first principal component accounts for as much of the variability in the data as possible , and each subsequent component accounts for the highest variance possible under the constraint that it is uncorrelated with preceding components [34 , 35] . For PCA , analyte concentrations were log transformed , and normalized to the mean . Associations between the PCs and CD4+-cell count , CSF lymphocyte count , baseline fungal burden , rate of clearance of infection , and macrophage activation marker concentrations were examined using Pearson’s correlation coefficient . Associations between PCs and rate of clearance of infection ( but not variables recorded at baseline ) were adjusted for treatment group . Differences in PC scores between patients who survived and died , and those who did and did not develop IRIS , were examined using a linear regression model adjusting for treatment group , CD4+-cell count , and the previously described risk factors for mortality , baseline fungal burden and altered mental status [36] . A sensitivity analysis exploring differences in PC scores between patients who survived and died , and those who did and did not develop IRIS , was preformed limited to the patients in the trial control arm who did not receive adjuvant interferon-γ immunotherapy . For detailed analysis exploring the relationships between individual cytokines , associations between continuous variables were explored using Pearson’s correlation coefficient and linear regression analysis . Variables were compared across groups using Student’s t-tests , Kruskal-Wallis tests , χ2 tests , or Fisher’s Exact tests as appropriate . Comparisons of paired groups were made using the Wilcoxon matched pairs test . Permutation tests , with 5000 permutations , were used to control the family wise-error rate ( FWER ) for the fifteen correlations / partial correlations between the cytokines IL-4 , IL-10 , IL-17 , MCP-1 and baseline and outcome variables . Permutation tests used the max statistic for adjusting the p-values to ensure FWER of 0 . 05 while allowing for the potentially strong correlations between variables . Unadjusted p-values for those associations that maintained a FWER<0 . 05 across the multiple comparisons are reported . Support vector machine ( SVM ) learning and decision tree analyses were used to assess the capacity for baseline CSF cytokine measurements to predict mortality or IRIS . The support vector machines algorithm was implemented in R v2 . 15 . 1 using the kernlab package and ksvm function with a linear kernel ( vanilladot ) , cost 1 and leave one out cross validation . Decision tree classification was performed using Waikato Environment for Knowledge Analysis ( WEKA ) v 3 . 6 . 9 using different decision tree classifiers ( Alternating Decisions Tree , Simple Cart , J48 Decision Tree and Random Forrest ) with leave one out cross validation . For the purposes of all analyses the two IFNγ treatment groups were considered as a single “IFNγ treated” group . Statistical significance was defined as p≤0 . 05 . The study was approved by the Human Research Ethics Committee of the University of Cape Town and St . George’s University of London . Patients gave written informed consent for blood and CSF samples to be used for research purposes .
Baseline CSF cytokine concentrations are shown in Fig 1 . IL-12 , IL-21 , IL-22 and IL-23 concentrations were below the limit of detection in the majority of cases , and have been excluded from subsequent analyses . Principal component analysis was used to identify co-correlated cytokine and chemokine measurements that accounted for the variance across the data set . The majority of the variance was reflected by PC1 ( 44% ) and PC2 ( 17% ) ( Fig 2A ) . The component loadings for each variable showed that the variance in PC1 was driven by positive loading scores for the pro-inflammatory cytokines IL-6 , IFNγ and the chemokine IL-8 ( Fig 2B ) . IL4 , IL10 and IL17 levels also made positive , albeit more modest contribution to the PC1 score , suggesting that Th1 ( IFNγ ) , Th2 ( IL4 and IL10 ) and Th17 ( IL17 ) responses were co-correlated in this context and confirmed by direct pairwise comparisons of each cytokine ( Fig 3 ) . The variance in PC2 was due to negative loading scores for the chemokines MCP-1 , MIP-1α , and the cytokine GM-CSF ( Fig 2B ) . Hence PC1scores were positively correlated with the levels of the pro-inflammatory cytokines IL-6 and IFNγ and PC2 scores were inversely correlated with the levels of the chemokines MCP-1 and MIP-1α ( Fig 2C and S2 Fig ) . PCA indicated that co-correlated pro-inflammatory cytokines and chemokines represented by PC1 scores , and those represented by PC2 scores function as independent variables . To assess whether proinflammatory cytokines or chemokines were related to other clinically important variables in cryptococcal meningitis we explored the relationship of PC1 and PC2 with peripheral blood CD4 , CSF lymphocyte counts , markers of CSF macrophage ( or resident CNS macrophage like cell ) activation , CSF arginase activity , baseline CSF fungal burden , rate of clearance of infection , 2-week mortality and IRIS ( Table 2 and Table 3 ) . PC1 scores showed a weak positive relationship with CD4 count and stronger relationship with CSF lymphocytes . PC2 scores correlated with both of these variables . These suggest that low levels of both peripheral CD4 cell count and CSF lymphocyte count were associated with reduced levels of pro-inflammatory cytokines but increased levels of the chemokines MCP-1 and MIP-1α in the CSF . Both PC1 and PC2 were also strongly positively correlated with the two soluble markers of macrophage activation in CSF , neopterin and sCD14 , however not with arginase activity ( Table 2 ) , indicating that macrophage activation was associated with robust pro-inflammatory CSF cytokine responses , and reduced chemokine expression . There were no significant associations between arginase activity and either IFNγ , IL-10 , or IL-4 . Fungal burden was negatively correlated with PC1 and PC2 indicating that patients with high fungal burdens tended to have pauci-inflammatory CSF cytokine profiles with high MCP-1 and MIP-1α chemokine expression , while lower fungal burdens were associated with more robust CSF inflammatory responses and lower chemokine levels . Importantly PC1 scores showed a positive correlation with more rapid clearance of cryptococci from the CSF ( an association that remained significant after adjustment for treatment group ) , consistent with the hypothesis that robust pro-inflammatory responses contribute to fungal clearance . Also in keeping with this observation , PC1 scores were lower in patients who had died within 2 weeks compared to those who survived ( Fig 4 ) . This association was significant following adjustment for treatment group , CD4 count , and the key predictors of mortality , baseline fungal burden and altered mental status ( p = 0 . 01 ) . There were no significant associations between PC2 and mortality ( Fig 4 ) . Amongst patients who survived , PC2 but not PC1 scores , differentiated patients who developed IRIS following the initiation of ART ( Fig 4 ) . PC2 scores in patients who developed IRIS were significantly lower than those who did not , showing that high chemokine expression at baseline , associated with low peripheral CD4 counts and CSF lymphocyte counts , was predictive of subsequent IRIS ( Fig 4 ) . The association between PC2 score and IRIS remained significant in analyses adjusted for treatment group , CD4+-cell count , and the previously described risk factors for mortality , baseline fungal burden and altered mental status , as described above ( p = 0 . 004 ) , and in a linear regression model adjusting for treatment group and the known risk factors for IRIS , baseline fungal burden and CSF white cell count ( p = 0 . 004 ) . The largest contributor to PC2 was MCP-1 , which was individually strongly negatively correlated with CD4 count ( r = -0 . 31 , p = 0 . 004 ) and log10 CSF lymphocyte count ( r = -0 . 46 , p<0 . 001 ) , and associated with IRIS ( baseline geometric mean CSF MCP-1 concentration 574 . 7 pg/ml in those not developing IRIS versus 2887 . 0 pg/ml in those who subsequently developed IRIS , p = 0 . 005 , Fig 5 ) . These associations all remained significant when controlling for a FWER of 0 . 05 . A sensitivity analysis examining associations between PC scores and outcomes restricted to the patients in the trial control arm who did not receive adjuvant interferon-γ immunotherapy was performed . In this smaller sample , the association between PC1 and rate of clearance of infection from the CSF and mortality were no longer significant ( early fungicidal activity Pearson’s-r = -0 . 12 , p = 0 . 5; mortality PC1 delta = -0 . 01 , p = 0 . 9 ) . The associations between PC2 and IRIS remained significant in both crude ( p = 0 . 01 ) and adjusted analysis ( p = 0 . 03 ) . Taken together , our data suggest that multiparameter measurements of CSF cytokines and chemokines in patients presenting with cryptococcal meningitis may predict the risk of early mortality or development of IRIS following ART . To test whether our data could be used to predict these clinical outcomes , we employed machine-learning algorithms and leave one out cross-validation to classify each patient after training on the data from all the other patients in the study . To take advantage of high dimensional pattern recognition we used support vector machine learning; and to assess whether a minimum number of variables could classify patients correctly , decision tree analyses was used ( Table 4 ) . Support vector machine learning ( SVM ) correctly predicted mortality or survival at two weeks in 84% of patients and IRIS in 83% of patients . Using a variety of decision tree analyses , we were also able to predict mortality or survival at two weeks with 80-84% accuracy and IRIS with 78-84% accuracy .
The phenotype of the baseline CSF immune response in patients with HIV-associated CM was shown to be associated with both microbiological and clinical outcomes , and predictive of subsequent IRIS . For the first time we have demonstrated that more robust pro-inflammatory CNS cytokine responses , characterized by higher levels of CSF IL-6 and IFNγ are associated with evidence of increased markers of innate effector cell activation , and , in keeping with previous reports [30] , more effective control of fungal burden , faster clearance of infection during treatment , and survival . These findings are consistent with animal model data demonstrating the importance of pro-inflammatory CNS cytokine responses [13 , 37 , 38] and microglial cell activation [38 , 39] in response to cytokines such as IFNγ for host resistance to cryptococcal infection . They also support previous findings showing that higher expression of pro-inflammatory cytokines such as IFNγ and TNFα by peripheral CD4+ T-cells [29] , and higher levels of IFNγ , TNFα , IL-6 , and IL-8 in the CSF of patients with HIV-associated CM [30] are associated with improved outcomes . In keeping with animal models [12 , 13 , 19 , 23 , 40] , our data suggests that the Th-17-type cytokine , IL-17 , plays an important role in the human CNS immune response to Cryptococcus . IL-17 concentrations were closely correlated with IFNγ levels , and associated with rate of clearance of infection and survival . Of note , IL-17 production was not observed in response to cryptococcal mannoprotein stimulation of CD4+ T-cells from a subset of patients included in this study[29] , suggesting that the source of CSF IL-17 may not be Th17-type CD4+ T-cells . Other cytokines associated with Th17-type T-cell responses , including IL-21 , IL-22 and IL-23 were not present in detectable quantities in the majority of patients studied , further supporting this supposition . γδ+ T-cells , CD8+ T-cells , NK cells , and neutrophils are all capable of producing the cytokine [21 , 40–44] , although it is not possible to determine the source of CSF IL-17 from our data . We found no evidence for a detrimental Th2-type response in CM patients , nor evidence for arginase activity , which in mouse models of Cryptococcal infection is associated with alternative activation of effector cells [16] . Levels of the classic Th2 cytokine IL-4 correlated closely with IFNγ concentrations , as did IL-10 concentrations , and both were associated with better control and clearance of cryptococcal infection and lower mortality ( although the mortality association with these individual cytokines did not remain statistically significant following adjustment for multiple comparisons , requiring validation in future studies ) . Th2-type T-cell responses , characterized by production of IL-4 and IL-13 and alternative activation of macrophages , are strongly associated with adverse outcomes in mouse models of cryptococcal infection [45] . Reasons for the differences between our findings and the animal models could relate to intrinsic differences between mouse and human immune systems , particularly in the context of advanced HIV-infection . The Th1/Th2 dichotomy described in mouse models appears less pronounced in humans [46] , and arginase production , which has been shown to be a marker of disease severity in HIV-infected patients [47] , is not well described in human macrophages , but rather is associated with myeloid derived suppressor cells and neutrophils [47] . It also almost certainly reflects the fact that the patients in our study had been infected with Cryptococcus for some time , and had had active CNS disease for several weeks prior to presentation [3] . Many of the mouse studies have examined local pulmonary responses during early infection , prior to dissemination of infection , and not CSF cytokine profiles . Additionally mouse models do not incorporate HIV-infection , or the effects of antifungal treatment . It is possible that a Th1/Th2 dichotomy exists during early infection in humans , but is not reflected in the CSF during the later stages of disease . Recent evidence suggests that both T-cell and macrophage polarization are dynamic and plastic processes [46 , 48] , and the CSF immune responses observed in the study patients likely represent a complex interaction of pro-inflammatory , compensatory anti-inflammatory and tissue repair processes that develop as disease evolves . In terms of cryptococcal IRIS , previous studies have identified poor baseline inflammatory responses , rapid immune reconstitution from this low baseline , and a high organism or antigen burdens as key risk factors [6] . Low peripheral CD4 counts and CSF lymphocyte counts , and reduced levels of IFNγ , TNFα , IL-6 , and IL-8 in the CSF at disease presentation have all been associated with increased risk of subsequent IRIS [7–9 , 49 , 50] . Although , in our study , patients who developed IRIS did tend to have a less inflammatory CSF cytokine response at baseline , the more striking finding was that development of cryptococcal IRIS following ART initiation was strongly associated with high CNS expression of the chemokines MCP-1 , MIP-1α , and the cytokine GM-CSF at initial CM presentation . This finding provides support for the findings of Chang et al . , who recently demonstrated that increased CNS expression of MCP-1 and MIP1α at baseline were associated with subsequent IRIS , and hypothesized that CD8+ T-cell and myeloid cell trafficking into the CNS in response to these chemokines predisposed to aberrant immune responses and excessive CNS inflammation following ART initiation [49] . In a previous study we have demonstrated that the phenotype of the peripheral CD4+ T-cell response to Cryptococcus is associated with disease severity and outcome in HIV-associated CM [29] . IFNγ and TNFα predominant responses were associated with increased CSF lymphocyte counts , higher levels of CSF cytokines including IL-17 , and survival . This study builds on these prior findings , suggesting that the presence of relatively small numbers of IFNγ and TNFα-producing Cryptococcus-specific memory CD4+ T-cells are capable of promoting a cellular infiltrate into the CNS in response to local production of chemokines such as MCP-1 and MIP1α by microglial cells following exposure to Cryptococci . We hypothesize that these infiltrating cells then stimulate expression of pro-inflammatory cytokines by both resident glial cells and trafficked cells of myeloid and lymphoid lineages , leading to activation of microglial cells and macrophages , with resultant restriction of intracellular growth of Cryptococci , resulting in lower fungal burdens at presentation , faster clearance on infection on treatment , and improved survival . With the evolution of this protective , predominantly Th1-type inflammatory cytokine response , compensatory parallel increases in both Th2-type cytokines such as IL-4 and regulatory cytokines such as IL-10 occur . In the absence of any effective Cryptococcus-specific CD4 response , microglial cells and CNS macrophages are unable to effectively restrict cryptococcal growth , leading to high organism burdens , poor clearance of infection during treatment , and high mortality . These patients with severe immune suppression at baseline , indicated by low peripheral CD4 counts and CSF lymphocyte counts , also have markedly up regulated CNS MCP-1 and MIP1α chemokine expression , which , following immune restoration with ART , leads to an influx of inflammatory cells , excessive local inflammation , and IRIS . Despite the prospective nature of our study , with comprehensive clinical and immunological data , there are a number of limitations . Firstly , the cellular source of the CNS cytokine production is not possible to elucidate from these data . Unlike in previous smaller studies [30 , 51] , levels of CSF cytokines were significantly correlated with both peripheral CD4+ cell count and CSF lymphocyte count . This may suggest that the source of these CSF cytokines is infiltrating T-cells and monocytes ( which may be mistaken for lymphocytes on CSF cell counts ) . Alternatively such incoming cells may facilitate or stimulate cytokine production by resident CNS immune cells . Interestingly a consistent and significant inverse relationship was seen between the levels of the chemo-attractant chemokines MCP-1 , MIP-1α and peripheral CD4+ cell count and CSF lymphocyte count . This finding is consistent with existing data suggesting that chemokine production by glial cells and monocytes/macrophages is not significantly compromised in patients with late stage HIV-infection [52] , and that the poor CNS inflammatory response observed is primarily the result of the absence of effective T-cell support . A biological feedback loop appears to be in operation such that chemokine production is rapidly down-regulated in the face of even minimal local lymphocyte infiltration and CD4+ T-cell help . Secondly , when exploring the correlations between baseline CSF immune parameters and fungal burdens it is not possible to definitively attribute causality . Whilst it seems plausible that the association between more vigorous CSF inflammatory responses and lower organism burdens means that variation in the patient’s ability to mount an effective inflammatory response plays a key role in determining disease severity , it is possible that very high organism burdens paradoxically lead to down regulation of the host inflammatory responses . Finally , these data were derived from a clinical trial comparing differing three treatment regimens leading to different rates of fungal clearance ( but not significant differences in mortality or rates of IRIS ) . The cytokine and chemokine levels used in these analyses were all determined prior to the administration of any antifungal therapy , meaning all baseline comparisons are unaffected by treatment group . And in all analyses relating to rate of clearance of infection , IRIS , and mortality outcomes , adjustment for treatment group was made . In each of these cases the associations seen in unadjusted analysis remained robust , or were strengthened following adjustment , supporting the validity of our findings . The sensitivity analysis restricted to the study control group ( i . e . those who did not receive adjuvant interferon-γ immunotherapy ) was underpowered as a result of the small sample size . However even in this small sample , although the mortality associations were no longer significant , the association between PC2 and IRIS remained statistically robust . In conclusion , the presence of a CSF inflammatory response consisting of Th1 , Th2 and Th17 type cytokines has been shown to correlate with markers of innate effector cell activation , more rapid clearance of infection and survival in patients with HIV-associated cryptococcal meningitis . We did not find evidence of a dichotomous Th1/Th2 response . Rather Th1 , Th2 and Th17 type cytokine concentrations correlate closely with each other , and the presence of this combined inflammatory response is beneficial , while its absence is detrimental . Patients with low peripheral CD4+ T-cell counts and CSF lymphocyte counts at disease presentation had high CNS chemokine expression , and were at high risk of subsequent IRIS following ART initiation . | Cryptococcal meningitis is a severe opportunistic infection , estimated to kill several hundred thousand HIV-infected individuals each year . One of the factors contributing to this high death toll is the inadequacy of antifungal treatments . As few novel antifungal drugs are being developed , several groups have started to investigate the potential of immune modulation , with treatments designed to change the patient’s immune response to infection . However , our understanding of the immune response to cryptococcal infection in HIV-infected patients , and how these responses impact on clinical outcomes , is limited . In this study , we took advantage of the fact that we can sample cerebrospinal fluid ( CSF ) from the site of the infection in patients when they develop cryptococcal meningitis . We undertook a detailed analysis measuring levels of immune response parameters in the CSF of these patients , and demonstrated that there were several distinct components of the immune response . Variations in these responses were associated with both the rate at which patients cleared their infection during treatment , and with mortality . Our results provide a basis for the development of future immunomodulatory therapies , and may allow identification of patients most at risk of dying , enabling more intensive treatments to be given to those at highest risk . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Cerebrospinal Fluid Cytokine Profiles Predict Risk of Early Mortality and Immune Reconstitution Inflammatory Syndrome in HIV-Associated Cryptococcal Meningitis |
The cysteine desulfurase IscS is a highly conserved master enzyme initiating sulfur transfer via persulfide to a range of acceptor proteins involved in Fe-S cluster assembly , tRNA modifications , and sulfur-containing cofactor biosynthesis . Several IscS-interacting partners including IscU , a scaffold for Fe-S cluster assembly; TusA , the first member of a sulfur relay leading to sulfur incorporation into the wobble uridine of several tRNAs; ThiI , involved in tRNA modification and thiamine biosynthesis; and rhodanese RhdA are sulfur acceptors . Other proteins , such as CyaY/frataxin and IscX , also bind to IscS , but their functional roles are not directly related to sulfur transfer . We have determined the crystal structures of IscS-IscU and IscS-TusA complexes providing the first insight into their different modes of binding and the mechanism of sulfur transfer . Exhaustive mutational analysis of the IscS surface allowed us to map the binding sites of various partner proteins and to determine the functional and biochemical role of selected IscS and TusA residues . IscS interacts with its partners through an extensive surface area centered on the active site Cys328 . The structures indicate that the acceptor proteins approach Cys328 from different directions and suggest that the conformational plasticity of a long loop containing this cysteine is essential for the ability of IscS to transfer sulfur to multiple acceptor proteins . The sulfur acceptors can only bind to IscS one at a time , while frataxin and IscX can form a ternary complex with IscU and IscS . Our data support the role of frataxin as an iron donor for IscU to form the Fe-S clusters .
Sulfur is a critical element in all living cells , incorporated into proteins not only in the form of cysteine and methionine but also as iron-sulfur clusters , sulfur-containing cofactors and vitamins , and into RNA through a variety of modifications [1] , [2] . Delivery of sulfur for these various biosynthetic pathways is a complex process , involving successive transfers of sulfur as persulfide between multiple proteins , many of which are highly conserved across species . Three distinct systems have been identified for the assembly of iron-sulfur clusters: isc , nif , and suf ( reviewed in [1] , [3]–[5] ) . The isc ( iron-sulfur clusters ) system participates constitutively in general-purpose iron-sulfur cluster assembly and in transfer of sulfur to several cofactors and tRNAs . The nif ( nitrogen fixation ) system is involved in iron-sulfur cluster assembly required for the maturation of nitrogenase [6] , while the suf ( sulfur mobilization ) system plays a role during oxidative stress or iron starvation . The initial step in each system is performed by a specific cysteine desulfurase , IscS [7] , NifS [8] , or SufS ( previously CsdB , [9] ) , respectively , forming the initial persulfide . IscS is a highly conserved , widely distributed pyridoxal-5′-phosphate ( PLP ) -dependent enzyme [7] , [10] , with 60% sequence identity between the enzyme from Escherichia coli and its human homolog , NFS1 . It initiates intracellular sulfur trafficking , delivering the sulfur to several sulfur-accepting proteins such as IscU , ThiI , TusA , and MoaD/MoeB that commit the sulfur to different metabolic pathways , including iron-sulfur cluster assembly , thiamine and biotin synthesis , tRNA modifications , or molybdopterin biosynthesis [2] , [3] , [11] . IscU is the primary scaffold for assembly of Fe-S clusters [12] that are required by iron-sulfur proteins . In addition to these sulfur acceptors , IscS interacts with several other proteins , including CyaY , a bacterial homolog of human frataxin [13] , [14]; IscX , a possible adaptor protein whose exact function is as yet unknown [15] , [16]; and rhodanese RhdA [17] . Frataxin/CyaY has been postulated as an Fe chaperone [18] , an Fe donor for Fe-S cluster assembly [13] , [19] , [20] , or a regulator of Fe-S cluster formation [14] . The network of known IscS protein interactions is shown in Figure 1 . Thiolated nucleotides are found in several tRNAs . In E . coli and Salmonella enterica serovar Typhimurium , these are s4U8 , s2C32 , ms2i ( o ) 6A37 , and ( c ) mnm5s2U34 , which , with the exception of s4U8 , are located within the anticodon loop and are crucial for proper mRNA decoding [21] . The base thiolations are mediated by several acceptor proteins , falling into two distinct pathways [21] . In the iron-sulfur cluster independent pathway , direct transfer of sulfur from IscS to the acceptor ThiI leads to the s4U8 modification [22] , while transfer to TusA results in the ( c ) mnm5s2U34 modification [23] . ThiI also participates in thiamine biosynthesis [24] . The second pathway proceeds through the formation of an iron-sulfur cluster and is dependent on the IscU acceptor protein . The enzymes TtcA and MiaB accept sulfur from IscU [3] and are responsible for the s2C32 [25] and ms2i ( o ) 6A37 modification [26] , respectively . The unique tRNA thiolation pattern associated with sulfur transfer from IscS to TusA , IscU or ThiI provides a convenient readout system to assess the in vivo effects of IscS mutations on its interaction with these proteins . The proteins involved in sulfur utilization have been extensively studied both functionally and structurally . Structures of IscS [27] , the sulfur acceptor proteins TusA [28] , ThiI [29] , IscU [30] , [31] , rhodanese [32] , and the modulators human frataxin [33] , [34] and its bacterial homologue CyaY [35] , [36] , as well as IscX [16] , [37] have been determined by X-ray crystallography or NMR . All of these proteins adopt different folds and the acceptor proteins receive sulfur from IscS by molecular mechanisms that are not fully understood . Despite this wealth of structural information , the question of how IscS is able to communicate with such a broad spectrum of proteins and deliver sulfur to a wide range of structurally divergent partners is unresolved as no structural information on its complex ( es ) with binding partner ( s ) is presently known . To begin addressing this question , we have determined the crystal structure of the IscS-TusA and the IscS-IscU complexes , which reveal different modes of binding of these proteins and provide a framework for understanding sulfur transfer from IscS . Further , we performed extensive mutagenesis of the IscS surface followed by in vitro ( pull-down ) and in vivo ( tRNA complementation assay ) studies to map the interface with ThiI , CyaY/frataxin and IscX . Competition for binding to IscS by its various partners has been explored by three-way pull-down experiments .
We have crystallized and determined the structures of the E . coli IscS-TusA and IscS-IscU complexes at 2 . 45 Å and 3 . 0 Å resolution , respectively ( Figure 2 and Table 1 ) . The atomic structures of these complexes provide a detailed description of two different protein binding sites on the IscS surface . IscS is composed of two domains [27] . The small domain ( residues 1–15 and 264–404 ) contains the critical active site cysteine Cys328 . The large domain ( residues 16–263 ) harbours the PLP cofactor and the cysteine substrate-binding pocket . Dimerization of IscS predominantly involves residues from the large domain . Easily recognizable electron density in our structures indicated the presence of the PLP cofactor as an internal aldimine covalently bound to Lys206 , as previously observed [27] . TusA has a compact two-layered α/β-sandwich structure with a central four-stranded mixed β-sheet having the connectivity β1↑β2↑β4↓β3↑ and two α-helices [28] . IscU is a two-layered α/β sandwich with a core three-stranded β-sheet and bundle of five α-helices [31] . The IscS-TusA complex crystallized in two forms with identical heterotetramers consisting of an IscS dimer and two TusA molecules . The distance between the two TusA monomers exceeds 40 Å ( Figure 2 ) . TusA interacts with the large domain of one IscS subunit within the dimer , with the exception of the tip of the loop containing the essential Cys328 of IscS , which comes from the other subunit ( Figure 2 ) . This persulfide-carrying Cys328IscS is juxtaposed against the acceptor cysteine of TusA , Cys19TusA , with only ∼4 Å separating their S atoms . Most of the IscS residues involved in the interaction with TusA are located on the outside face of a six-turn helix α2 , the N-terminus of strand β2 , the C-terminus of the neighbouring strand β9 , and the following loop β9/α7 ( Figure 3A and Figure S1 ) . Electron density for the interface residues is shown in Figure S2A . The residues of TusA contacting IscS are located on two α-helices ( α1TusA and α2TusA ) , which are nearly perpendicular to helix α2IscS . Formation of the complex buries the α-helical layer of TusA and leaves its β-sheet layer exposed to the solvent . Approximately 710 Å2 of the molecular surface of each binding partner is buried , corresponding to ∼16% of the total TusA surface area . The interface involves van der Waals contacts , polar and hydrogen bond interactions , and salt bridges ( Figure 3A ) . The main van der Waals contacts are provided by TusA Met24TusA , Met25TusA ( α1TusA ) , Phe55TusA , Phe58TusA , Met59TusA ( α2TusA ) and IscS Trp45IscS ( stacking with Phe58TusA ) , and the aliphatic portions of Arg55IscS and Arg237IscS . As established previously [38] , the IscS-IscU complex is also a heterotetramer . IscU binds near the C-terminus of IscS , forming a very elongated S-shaped heterotetrameric protein complex 150 Å long and 65 Å wide ( Figure 2 ) . The IscU is in its apo form , with no evidence of a bound Fe-S cluster . IscU makes contacts with helix α8IscS ( Glu309-Ala316 ) , helical turn α10IscS ( Glu347 ) , the end of helix α11IscS , and the C-terminal helix α12IscS ( Arg379-Lys391 ) . The importance of the latter contact is emphasized by the lack of binding of IscU to IscS ( Δ376-404 ) [39] . The contacts on IscU include Tyr3 and Tyr11 ( N-terminus ) , Gly38 , Val40 and Lys42 ( β2IscU ) , Lys59-Gly64 ( β3IscU ) , and Lys103 ( Figure 3B , electron density in Figure S2B ) . The IscU surface area buried upon complex formation is ∼790 Å2 . The bound IscU projects its most conserved surface containing three conserved cysteines ( Figure S3 ) toward the IscS loop that carries Cys328 . The distance between the modeled Cys328IscS and any cysteine of IscU in our structure is greater than ∼12 Å , implying that a conformational change must accompany sulfur transfer ( Figure S4 ) . The contacts provided by the N-terminus and helix α1IscU ( Glu5-Glu12 ) are critical for the formation of the cognate complex , as confirmed by a partial loss of in vitro binding of IscU ( Δ1-7 ) to IscS and a complete loss of binding of IscU ( Δ1-12 ) ( Table 2 and Figure S5A ) . We constructed several IscU point mutants of residues on loops facing IscS to verify the interface observed in the IscS-IscU structure . Only the charge reversal mutant K103IscUE located within the interface and pointing toward IscS disrupted the complex ( Table 2 ) . Removing the sidechain of another residue located at the interface , Tyr11 ( Y11A ) , had no significant effect on binding as this was not a disruptive mutation . Finally , the charge removal/reversal mutants E5L , D9R , and E98R located outside the observed interface had no effect on complex formation . To determine if the IscS-TusA and IscS-IscU complexes existing in solution are the same as the heterotetramers observed in the crystal structures , we performed small angle X-ray scattering ( SAXS ) experiments . The scattering curve obtained for the IscS-TusA complex at a protein concentration of 22 mg/ml fit very well ( χ2 = 2 . 24 ) to the intensity profile calculated from the crystal structure of the complex ( Figure 4 ) , indicating that the crystal and solution structures represent the same biological unit . Similarly , the data for the IscS-IscU complex are in excellent agreement ( χ2 = 1 . 22 ) with the very elongated structure observed in the crystal ( Figure 4 ) . Formation of the IscS-TusA or IscS-IscU complexes is associated with only minor conformational changes in the IscS dimer , predominantly of surface sidechains . The root-mean-square deviation ( rmsd ) between free ( PDB code 1P3W ) and TusA-bound IscS is ∼0 . 4 Å for the corresponding ∼380 Cα atoms . Nevertheless , sidechain reorientation results in a significant change in the shape of the IscS binding surface and improves surface complementarity to TusA ( Figure 5 ) . There is no change in the active site pocket containing the PLP cofactor . The TusA molecules in the complex show larger structural deviations from the individual TusA structures as determined by NMR spectroscopy ( PDB code 1DCJ , [28] ) ( rmsd of ∼1 . 3 Å for all Cα atoms ) , corresponding to a ∼2 . 5 Å shift of helix α2TusA away from α1TusA along the surface of the β-sheet , accompanied by a small ∼15° rotation of this helix along its axis . Upon binding of IscU to IscS , the major structural change in IscU relative to the solution structures of IscU from H . influenza [30] , B . subtilis ( PDB code 1XJS ) , and mouse ( PDB code 1WFZ ) involves ordering of the ∼25 N-terminal residues and folding of Glu5-Glu12 into an α-helix , thereby providing crucial contacts with IscS . This segment is largely disordered in all solution structures of IscU and the N-terminus assumes different conformations in three independent molecules in the crystal structure of Aquifex aeolicus IscU [31] . The rmsd between E . coli IscU and Aquifex aeolicus IscU is ∼1 . 3–1 . 6 Å for the ordered ∼100 Cα atoms segment . The structures of IscS-IscU and IscS-TusA identified non-overlapping IscS surfaces ( with the potential exception of the disordered tip of Cys328IscS loop ) interacting with IscU and TusA . However , IscS also interacts with several other proteins and we aimed to identify the “active” surface of IscS . We first analyzed the pattern of surface residue conservation using the CONSURF server ( http://consurf . tau . ac . il/; [40] ) . The conserved residues form a large , contiguous molecular surface extending across the dimer interface and centered on the active site Cys328 ( Figure 6A ) . The extent of the conserved surface suggests that a substantially larger surface area than that observed for the IscS-IscU and IscS-TusA complexes is utilized for binding all protein partners . To further characterize the IscS binding surface we expressed and purified three other proteins in addition to IscU and TusA , namely the sulfur acceptor ThiI , a modulator frataxin/CyaY , and IscX from the isc operon . All of these proteins have previously been shown to bind to IscS . The IscS utilized in this study had not been charged with the persulfide group . Nevertheless , all IscS partners formed stable complexes , indicating that Cys328 does not need to be present in the persulfide form for protein-protein binding ( see below ) . To experimentally map the IscS interacting surface , we created a series of IscS point mutations distributed across the entire conserved surface ( Figure 6A and Table 3 ) . The mutations were designed to invert the polar or nonpolar character of a specific residue , or replace a smaller sidechain by a larger one . For in vitro pull-down experiments , all mutant proteins were expressed and purified following the same protocol as for wild-type IscS and showed similar behaviour during purification . IscS mutations that abrogated interaction with wild-type TusA , W45IscSR , E49IscSA , D52IscSR ( Figure 7A ) , D52IscSY , and D52IscSM ( unpublished data ) involved tightly clustered residues located on the side of helix α2IscS , in excellent agreement with the crystal structure . A significant contribution of hydrophilic interactions to IscS-TusA complex formation was demonstrated by disruption of the complex through increasing the NaCl concentration to 600 mM ( unpublished data ) . Of the IscS mutations , only A327IscSV had some impact on IscU binding ( Table 3 and Figure 7B ) . This mutation affects the residue next to Cys328 , and the tip of this loop was disordered in our structure . No other IscS mutations investigated here affected IscS-IscU complex formation and the structure shows that all of these mutations are outside of the IscU interface with IscS ( Figure 6B ) . However , an IscS ( Δ374-404 ) deletion was reported to abrogate IscU binding [39] , and this segment forms part of the interface observed in the structure . The agreement between the pull-down experiments and the crystallographically determined interfaces substantiated the results presented below for other proteins interacting with IscS . E . coli ThiI is significantly larger than either TusA or IscU , with 482 residues arranged into three domains [29] . The ThiI residue Cys456 was shown to be essential for accepting sulfur from IscS [41] , [42] and is located in the rhodanese-like domain . The mutants R220IscSE , R237IscSE/M239IscSE , and R340IscSE significantly decreased binding of ThiI , while the mutations W45IscSR , F89IscSE , R116IscSE , R223IscSE , E311IscSR , and A327IscSV decreased binding to a lesser extent ( Figure 7C and Table 3 ) . Therefore , binding of TusA or ThiI to IscS is influenced by a common mutation , W45IscSR , indicating that they bind to distinct but partially overlapping regions on the IscS surface . The binding of frataxin/CyaY and IscX to IscS was affected by the same set of mutations , including R116IscSE , R220IscSE , R223IscSE , R225IscSE/E227IscSR , G234IscSL , R237IscSE/M239IscSE , A327IscSV , and R340IscSE ( Figure 7D , E and Table 3 ) , showing that their footprints are very similar . Moreover , their footprints overlap significantly with that of ThiI but not with that of IscU nor TusA . The effect of IscS mutations on binding to partner proteins was analyzed in vivo by quantification of the tRNA modifications mnm5s2U ( TusA ) , s2C ( IscU ) , and s4U ( ThiI ) . To this end , we used an iscS null mutant ( IC6087 ) transformed with pMJ623 and derivative plasmids , which encode the wild-type and mutant His-IscS proteins , respectively . We decided to use this approach after observing that plasmid pMJ623 was able to restore the nearly wild-type levels ( 90% ) of thiolated nucleosides when transformed into IC6087 , despite that His-IscS could not be detected with anti-His antibody in Western blot analysis ( unpublished data ) . Mutations W45IscSR , E49IscSA , D52IscSA , D52IscSR , D52IscSY , and D52IscSM reduce the mnm5s2U synthesis to 0%–25% of the wild-type protein , whereas they do not affect s2C accumulation . These results correlate well with the effect produced by such mutations on the IscS interaction with TusA and IscU , as assessed by the pull-down experiments ( Table 3 ) , suggesting that the impairment or complete inability of IscS mutants to bind TusA is responsible for the decrease in mnm5s2U modification . The mutation A327IscSV does not interfere with the pull-down of IscS by TusA , although it reduces the mnm5s2U synthesis by about 50% [21] , [43] . The mutation W45IscSR decreases both mnm5s2U and s4U levels to about 5% of the wild-type protein , confirming that Trp45 affects binding to TusA and ThiI ( Table 3 and [21] ) . However , other mutations impairing the interaction with TusA ( E49IscSA , D52IscSA , D52IscSY , and D52IscSM ) do not reduce synthesis of s4U , suggesting that they do not abrogate the interaction with ThiI . These results support that TusA and ThiI bind to distinct but partially overlapping regions on the IscS surface . Taken together with the determined structures , the in vitro and in vivo experiments enabled us to create a protein interaction map of the IscS surface ( Figure 6B ) . Structures of the IscS-TusA and IscS-IscU complexes showed that the footprints of TusA and IscU on the IscS surface do not intersect . Therefore , we applied a three-way pull-down approach to explore whether both of these proteins could bind simultaneously to IscS . We first incubated His6-IscS with GST-TusA on glutathione Sepharose beads , washed the beads extensively , and eluted the His6-IscS-TusA complex by cleavage with TEV protease . We then bound GST-IscU on fresh glutathione Sepharose beads , washed , and added the His6-IscS-TusA complex . The column was washed , TEV protease added , and incubated for ∼2 h . Only His-IscS and IscU eluted from the column ( Figure S6A , left ) . In the second experiment , we first formed the His6-IscS-IscU complex and loaded it on a glutathione Sepharose column pre-bound with GST-TusA . In the flowthrough we detected His-IscS-IscU . All of the GST-TusA and a small amount of His-IscS were retained on the beads ( Figure S6A , right ) . In both experiments IscS associated predominantly with IscU , indicating that TusA and IscU cannot bind to IscS simultaneously and that IscU is able to displace TusA from IscS . The biological significance of this binding preference has to be investigated further . Subsequently , we performed three-way pull-down experiments for other protein-protein combinations with IscS , including IscU-CyaY ( Figure S6B ) [13] , [14] , IscU-IscX ( Figure S6C ) , TusA-IscX ( Figure S6D ) , and TusA-CyaY ( Figure S6E ) . The results show that IscU can bind IscS simultaneously with either CyaY or IscX , whereas TusA cannot . To determine if simultaneous binding of CyaY ( or IscX ) and IscU to IscS affects sulfur transfer to IscU , we examined the level of IscU-dependent s2C tRNA modification when CyaY ( or IscX ) was overexpressed for 18 h . No effects were found ( unpublished data ) . As previously observed , both CyaY and IscX contain a large , negatively charged patch on their surface that has been proposed to contain residues involved in binding to IscS [14] , [16] , [37] . The CyaY and IscX footprints on the IscS surface encompass a positively charged area ( Figure 8 ) . We have used the ZDOCK server ( http://zdock . bu . edu/ ) to model the IscS-CyaY and IscS-IscX complexes . In the first approach no restraints were provided . While the 20 top solutions positioned CyaY over the positively charged surface of IscS near the Cys328 loop , the orientation of CyaY varied significantly and all the top solutions collided with IscU . In the second approach we provided CyaY residues identified by NMR [14] as restraints . Again , more than half of the 20 best models collided with IscU . However , when we added IscS restraints derived from pull-down assays , none of the top 20 solutions clashed with IscU , and the range of CyaY orientations was smaller than in the previous calculations ( Figure 9 and Figure S7 ) . What is more , all of the CyaY models collided , albeit slightly , with the TusA structure ( Figure 9 ) . This is consistent with the detection of an IscS-IscU-CyaY ternary complex and the lack of detection of an IscS-TusA-CyaY complex . Similar modeling results were obtained for IscX ( unpublished data ) . The crystal structures presented here allow us to address the mechanism of sulfur transfer from IscS to acceptor proteins . In the IscS-TusA complex , the observed proximity of Cys19TusA to persulfated Cys328IscS could be sufficient for sulfur transfer to occur . However , several residues , including Asp45TusA and Asp51TusA in the vicinity of Cys19TusA , are absolutely conserved and could play a role in sulfur transfer ( Figure 3A ) . Asp51TusA is on the surface while Asp45TusA is buried but forms a hydrogen bond to the NH of Cys19TusA . To investigate their roles , we constructed mutations D45TusAA and D51TusAA as well as other mutations affecting TusA residues in proximity to IscS , E21TusAA , M24TusAR , R27TusAE , R27TusAD , R31TusAA , and F58TusAA , and tested each mutant for IscS-TusA complex formation in vitro ( Figure S5B ) and in vivo for levels of TusA-dependent mnm5s2U tRNA modification ( Table 4 ) [23] . For the in vivo experiments we followed the synthesis of mnm5s2U in a tusA null mutant ( IC6085 ) transformed with pGEX 4T-1 ( expressing only GST ) and derivative plasmids expressing wild-type or mutant GST-TusA proteins . Western blot analysis with an anti-GST antibody indicated that the recombinant proteins are synthesized even in the absence of the IPTG inducer , due to leakiness of the Ptac promoter , and that the cellular levels of the GST-TusA protein produced by each recombinant plasmid under such conditions were similar ( unpublished data ) , suggesting that the introduced mutations did not affect stability of the GST-TusA protein . In all cases where the mutants show weak or no interaction in the pull-down assay , the level of tRNA modification also decreases ( Table 4 ) . Even when we detected no interaction by in vitro pull-downs , the remaining low IscS-TusA affinity seems to be sufficient to provide partial complementation over the several hours of cell growth , accounting for the reduced levels of tRNA modification observed in such cases ( Table 4 ) . The TusA interface mutations M24TusAR , R27TusAE , R27TusAD , R31TusAA , and F58TusAA abolished in vitro binding to IscS , while E21TusAA only weakened complex formation with IscS ( Table 4 ) . A more sensitive technique , surface plasmon resonance ( SPR ) , did not detect interaction between His-IscS and several of these TusA mutants ( M24R , R27E , R31A , F58A ) ( unpublished data ) . On the other hand , the D51TusAA and D45TusAA mutants behaved like wild-type TusA in the pull-down experiments with IscS , showing that these mutations had little or no effect on IscS-TusA complex formation ( Table 4 ) . When assayed in vivo , D51TusAA and D45TusAA showed reduced levels of mnm5s2U modification , to 67% and 56% , respectively , of that of the wild-type TusA ( Table 4 ) , supporting a functional role for Asp45 and Asp51 .
IscS and several of its binding partners are evolutionarily highly conserved proteins . In order to characterize at the molecular level the mode of interaction of IscS with its binding partners and to define their footprints on the IscS surface , we determined the crystal structures of IscS with two sulfur acceptors , IscU and TusA . We also utilized data from the literature for 9 mutations [21] , [39] , [44] with over 20 mutations investigated here to map interactions for three other proteins , ThiI , CyaY/frataxin , and IscX . We identified multiple mutations that disrupted binding for each of the partners ( Table 3 ) . The in vivo effects largely coincide with the in vitro binding studies ( Table 3 ) , offering supporting evidence that disrupting the interactions of IscS with its partners impairs tRNA modification . The structures of the IscS-TusA and IscS-IscU complexes validated this methodology . The footprints of ThiI , CyaY , and IscX overlap significantly , while ThiI and TusA overlap partially ( Figure 6B ) . Our results indicate that CyaY and IscX bind to nearly the same region of IscS . Although the TusA and IscU footprints do not overlap , the three-way pull-down experiments showed that TusA and IscU cannot bind simultaneously to IscS . Moreover , IscU was able to displace TusA in the complex , suggesting that it has a higher affinity for IscS . Superposition of the structures of these two IscS complexes shows , indeed , a spatial overlap between bound IscU and TusA ( Figure S4 ) . Taken together , our data show that the sulfur acceptors IscU and TusA and ThiI can bind to IscS only one at a time and that the effectors/modulators CyaY/frataxin and IscX can form a ternary complex with IscS in the presence of IscU but not with TusA or ThiI . As CyaY and IscU can both bind to IscS simultaneously , we asked if CyaY may prevent IscU from acquiring sulfur from IscS in vivo . To determine this we overexpressed CyaY or IscX in a wild-type E . coli strain and quantified the level of the modified s2C nucleotide , finding that overexpression has no effect on s2C synthesis under our growth conditions ( unpublished data ) . Several , and often contradictory , views on the role of frataxins have been proposed . Thus , Frataxin/CyaY has been postulated as an Fe chaperone [18] , an Fe donor for Fe-S cluster assembly [13] , [19] , [20] , or a regulator of Fe-S cluster formation [14] . Since we did not detect impairment in s2C modification under CyaY overproducing conditions , it may be concluded that CyaY does not interfere with sulfur transfer between IscS and IscU under standard growth conditions , which favours the view of CyaY as a source of Fe via IscU for Fe-S cluster assembly . However , some biochemical studies on frataxins suggest that their activity might be modulated in vivo by the intracellular iron concentration [14] or redox potential [19] . Therefore , additional experiments are needed to test the effect of the CyaY overexpression under such conditions . Each IscU molecule interacts with only one subunit of the IscS dimer and , based on its orientation in the complex , would be expected to accept sulfur from the same subunit to which it is bound ( Figure 2 ) . Of the three cysteines in IscU , the closest to the loop bearing Cys328IscS is Cys37IscU . The tip of the IscS loop is disordered and we cannot precisely position Cys328IscS , however the distance of ∼12 Å estimated from the model would be too far for sulfur transfer . The other two cysteines are slightly further away , with distances of ∼13 . 5 Å for Cys63IscU and ∼16 Å for Cys106IscU . Therefore , an additional movement , most likely of the IscS loop , is required to bring the catalytic Cys residues closer together . The mode of TusA interaction with IscS is different . While TusA interacts predominantly with one IscS subunit , the sulfur accepting Cys19TusA [23] is juxtaposed against Cys328′IscS that belongs to the other IscS subunit of the dimer ( Figures 2 , 3A ) . As a result , the thiol groups of Cys328′IscS and Cys19TusA are in close proximity , within a distance of less than 4 . 5 Å . This organization of the IscS-TusA complex suggests that the dimerization of IscS is essential for effecting sulfur transfer to various acceptor proteins . While the catalytic mechanism of cysteine and selenocysteine desulfurase/deselenase activity has been intensively investigated [45]–[48] , less is known about how persulfide sulfur is transferred to an acceptor protein . Evidence suggests that the cysteine persulfide intermediate is a relatively stable species and represents a true enzyme intermediate along the reaction pathway [1] , [49] . The loop containing Cys328IscS , which would carry the persulfide , extends away from the PLP cofactor and the cysteine-binding site , but the location of its tip harbouring Cys328IscS could not be detected due to disorder [27] . We have determined the structure of PLP-bound IscS at 2 . 05 Å resolution in a different crystal environment from that observed previously and have also found the Cys328IscS-containing loop extending away from the protein with its tip disordered . Therefore , IscS prefers an “open” conformation of the Cys328 loop , compatible with sulfur transfer to an acceptor . In contrast , the analogous loops in two other cysteine desulfurases , NifS and SufS , are shorter and prefer a closed conformation , with the active site cysteine residue located in proximity to PLP , compatible with loading of sulfur acquired from bound cysteine substrate ( Figure S8 ) . We postulate that the longer Cys328 loop found in IscS is essential for this enzyme to transfer sulfur to multiple acceptors . We propose that the transfer of persulfide sulfur from IscS to the acceptor occurs in two stages . In the first stage , the loop containing Cys328 assumes the “closed” conformation and is loaded with the sulfur acquired from the cysteine substrate via the PLP cofactor , as exemplified by the structure of SufS/CsdB [49] . Next , the Cys328-carrying loop pivots around hinges located near Ser324 and Ser336 , adopting the “open” conformation such that Cys328 can closely approach the cysteine of the acceptor protein . The conformation of the Cys328IscS loop in the IscS-TusA complex , with the donor and acceptor cysteines in close proximity , suggests that the observed conformation is close to that expected in a transfer-competent state ( Figure 3A ) . This transfer mechanism is likely common with both NifS and SufS desulfurases . IscS transfers sulfur to multiple acceptor proteins . In the complex with IscU the observed distance between Cys328IscS and the Cys residues of IscU is too long for a direct transfer ( Figure S4 ) , and consequently a conformational rearrangement is necessary to bring together the sulfur donor and acceptor cysteines . Since most regions of IscS show no differences in the various crystal structures , either alone or complexed with acceptor proteins , and in view of the high flexibility/disorder of the Cys328 loop , we postulate that it is this loop that bends closer toward IscU in order to effect sulfur transfer . Indeed , the observation of a disulfide linkage between Cys328IscS and Cys37IscU from Azotobacter vivendi [50] or with E . coli Cys63IscU [51] supports the notion that the Cys328 loop travels over a significant distance in order to interact with different partners . This implies that the flexibility of the Cys328 loop is crucial for the IscS ability to act as a shuttle in sulfur transfer and is consistent with the in vivo effects of mutations in the loop region of IscS on Fe-S cluster synthesis [21] , [43] , [44] . Our observation that the A327IscSV mutation weakens the IscS interaction with IscU , ThiI , CyaY , and IscX is also compatible with this hypothesis . Given that Ala327IscS is adjacent to the catalytic Cys328IscS , the mutation A327IscSV likely affects the flexibility of the active loop , resulting in impaired binding of IscS to some of its partners . The modeled position of Cys328IscS is closer to Cys37IscU and Cys63IscU than to Cys106IscU . The sidechain of Cys37IscU is exposed on the protein surface , with Cys63IscU being less exposed while Cys106IscU is buried . We propose that the most likely candidate residue to act as the initial S acceptor is Cys37IscU followed by Cys63IscU . The distance between the sidechains of Cys63IscU and Cys106IscU is ∼4 Å , allowing for a secondary transfer of persulfide sulfur from Cys63IscU to Cys106IscU . The observation that mutation of any one of the IscU cysteines reduced the number of sulfurs bound to IscU but did not abolish sulfur transfer [50] indicates that more than one cysteine can accept the sulfur directly from IscS . We questioned if sulfur transfer between the two cysteines requires assistance from other residues . We noted that Asp45TusA and Asp51TusA are close to Cys19TusA ( Figure 3A ) and are conserved in all homologs with sequence identity > ∼24% . The sidechain of Asp45TusA forms a hydrogen bond to the NH of Cys19TusA that may be helpful to correctly orient the loop carrying this cysteine . The sidechain of Asp51TusA is 4 . 2 Å away from the sulfur of Cys19TusA . The expected chemistry requires that Cys19TusA acts as a nucleophile attacking the Cys328IscS persulfide , for which Cys19TusA would be more reactive if it were deprotonated [1] . While at neutral pH a small fraction of cysteines would be deprotonated , we rationalized that Asp51TusA could act as a general base to deprotonate Cys19TusA . The D51TusAA mutation modestly affects sulfur transfer , as measured by the level of mnm5s2U modification in vivo , whereas , as expected , it does not impede IscS-TusA complex formation in vitro ( Table 4 ) . Therefore , we postulate that while Asp51TusA is not absolutely essential , it makes Cys19TusA more nucleophilic , increasing the enzyme's efficiency and resistance to changes in pH . The sulfuryl anion would also be stabilized by the nearby Arg50TusA . This residue , while only moderately conserved , also has a functional role as the R50TusAA mutant shows reduced tRNA modification without affecting IscS-TusA complex formation ( Table 4 ) . While our proposal is in agreement with the current data , more detailed investigations of the sulfur transfer reaction in vitro will be needed to establish the roles of the above-mentioned residues . Interestingly , an aspartate ( Asp39IscU ) has also been shown to destabilize the Fe-S cluster in IscU [31] , [38] . Mutation of this aspartate to an alanine was essential for crystallization of the Aquifex aeolicus IscU- ( Fe-S ) 2 cluster . This aspartate is located in between Cys37IscU , Cys63IscU , and Cys106IscU and we hypothesize that , by analogy to Asp51TusA , it could also participate in catalysis . Our combined biochemical and structural studies provide the first molecular details of how IscS both recognizes and discriminates between various binding partners . IscS binds its partners via a large , highly conserved , contiguous docking surface extending across both IscS subunits and centered on the loop containing Cys328 . Different binding partners utilize different parts of this docking surface and approach Cys328 from different directions . The key to the ability of IscS to transfer persulfide sulfur to multiple acceptor proteins is the length and flexibility of the loop carrying Cys328 . Indeed , superposition of the complexes shows that Cys19TusA and Cys37IscU are over 16 Å apart ( Figure S4 ) , yet both can accept sulfur from Cys328IscS . The shorter loops carrying the active site cysteine in SufS and NifS are likely adapted for interaction with only a single acceptor protein , SufU and NifU , respectively , and may require the binding of this partner to trigger flipping of this loop from an inside conformation to an outside one . It is clear that IscS binds the monomeric form of apo-IscU , consistent with the model proposed by Shimomura et al . [31] , and would be structurally inconsistent with binding of an IscU trimer containing an Fe-S cluster . It is also noteworthy that the binding site on IscU for the HscA chaperone , required for Fe-S cluster assembly or delivery from IscU to target proteins , may have some overlap with that for IscS since Lys103IscU was shown to be involved in HscA binding [52] and the K103IscUE mutation also disrupts the IscS-IscU complex ( Figure S5A and Table 2 ) . This argues against simultaneous binding of IscU to IscS and HscA and is consistent with a role for this chaperone in mediating delivery of the Fe-S cluster to recipient proteins . On the other hand , the IscU binding site for the co-chaperone HscB [53] is distinct from that for IscS and HscB could interact with the IscU-IscS complex . Since formation of an Fe-S cluster likely occurs while IscU is bound to IscS [38] and HscA affinity for IscU increases ∼20-fold in the presence of HscB [54] , a plausible model is that HscB promotes dissociation of the IscS-IscU ( Fe-S ) complex and a formation of an IscU ( Fe-S ) -HscB-HscA complex for subsequent transfer of the Fe-S cluster to a recipient protein . Within the cell , the relative affinities of partner proteins for the IscS dimer , their Fe-loading state ( IscU , CyaY , and IscX ) , as well as their relative concentrations together presumably dictate which combination ( s ) of partner proteins interact with IscS at any one time . The simultaneous binding of TusA and IscU to IscS , while it involves different surface residues on IscS , is precluded due to steric clashes . The higher affinity of IscS for IscU than for TusA suggested by our results is of functional importance in that under conditions of limited sulfur supply , sulfur would be delivered predominantly to IscU , the precursor for Fe-S cluster assembly . The overlapping footprints of ThiI and TusA on the IscS surface suggests that they cannot bind IscS simultaneously and , therefore , implies that synthesis of modified tRNAs containing S4U and S2U depends on binding competition between these two proteins . The pertinent question of the precise order of events at the molecular level leading to Fe-S cluster assembly on IscU , with respect to donation of Fe and S atoms , remains an area for further research .
The iscS gene ( NCBI gi: 12516934 ) from E . coli O157:H7 EDL933 [55] was cloned into a modified pET15b vector ( Novagen ) and was expressed in E . coli BL21 ( DE3 ) , yielding a fusion protein with an N-terminal His6-tag . The tusA ( NCBI gi:12518129 ) , iscU ( gi:12516933 ) , thiI ( gi:26106827 ) , iscX ( gi:12516925 ) , and cyaY ( gi:12518674 ) genes from the same bacterium were cloned into a modified pGEX-4T1 vector ( GE Healthcare , Baie d'Urfe , Quebec , Canada ) and expressed in E . coli BL21 as N-terminal glutathione S-transferase ( GST ) fusion proteins with a tobacco etch virus ( TEV ) protease cleavage site for removal of the tag . For each protein , an overnight culture of transformed E . coli BL21 was used to inoculate a 11 culture in TB medium containing 100 µg/ml ampicillin . The culture was grown at 37°C until the absorbance at 600 nm reached 0 . 6 . Protein expression was induced with 100 µM isopropyl 1-thio-β-D-galactopyranoside ( IPTG ) followed by incubation for 16–20 h at 20°C . Cells were harvested by centrifugation ( 4 , 000×g , 4°C , 25 min ) and stored at −20°C . The cell pellet was re-suspended in 40 ml of lysis buffer ( 50 mM Tris-HCl pH 8 . 0 , 0 . 15 M NaCl , 5% ( v/v ) glycerol ) . To obtain the IscS-TusA complex , the cell pellets of His6-IscS and GST-TusA were mixed and disrupted by sonication ( 12×10 s , with 10 s between bursts ) . Cell debris was removed by centrifugation ( 33 , 000×g , 45 min , 4°C ) . The protein supernatant was loaded onto a 2 ml bed volume of glutathione Sepharose resin ( GE Healthcare , Mississauga , Canada ) equilibrated with lysis buffer . Beads were washed with 4 column volumes of TEV cleavage buffer ( 50 mM Tris pH 8 . 0 , 150 mM NaCl , 0 . 5 mM EDTA ) to remove unbound proteins and excess IscS protein . The complex was released from the column by cleavage with TEV protease ( 1∶100 [wt/wt] ) for 3 h at room temperature . The IscS-TusA complex was further purified by size exclusion chromatography ( SEC ) on a Hi-Load Superdex 200 16/60 column ( GE Healthcare ) equilibrated in a buffer containing 20 mM Tris-HCl pH 8 , 150 mM NaCl , 2% ( v/v ) glycerol . Fractions containing the protein complex were pooled and concentrated to 35 mg/ml . The IscS-IscU complex was purified in a similar manner . Dynamic light scattering measurements were performed at room temperature using a DynaPro plate reader ( Wyatt Technologies , Santa Barbara , CA ) . Initial crystallization conditions were found by sitting drop vapour diffusion at 21°C using Qiagen JCSG Core Suite screens ( Qiagen , Mississauga , Canada ) and optimized by hanging drop vapour diffusion methods . The best crystals of IscS-TusA were grown by equilibrating 1 µl of protein ( 35 mg/ml ) in buffer ( 20 mM Tris-HCl pH 8 , 150 mM NaCl , 2% ( v/v ) glycerol ) mixed with 1 µl of reservoir solution ( 0 . 12 M magnesium formate , 20% [w/v] PEG 3350 ) suspended over 1 ml of reservoir solution . Two crystal forms were obtained under the same crystallization conditions . Crystals for form 1 are orthorhombic , space group P212121 , with a = 72 . 3 , b = 106 . 5 , c = 122 . 1 Å , with an IscS dimer and two TusA molecules in the asymmetric unit and Vm = 2 . 05 Å3 Da–1 [56] . Crystals of form 2 are also orthorhombic , space group C2221 , with a = 72 . 9 , b = 131 . 4 , c = 106 . 4 Å , with one IscS subunit and one TusA molecule in the asymmetric unit and Vm = 2 . 28 Å3 Da–1 . Crystals of IscS-PLP were obtained from 0 . 1 M Bicine pH 8 . 5 , 15% ( w/v ) PEG 6000 and belong to space group P212121 , with a = 74 . 8 , b = 99 . 2 , c = 118 . 1 Å , and Vm = 2 . 43 Å3 Da–1 . The best crystals of the IscS-IscU complex were obtained by hanging drop vapour diffusion by mixing 1 µl of IscS-IscU ( 30 mg/ml ) in buffer ( 20 mM Tris-HCl pH 8 , 100 mM NaCl , 2% v/v glycerol ) with 1 µl of reservoir solution ( 0 . 2 M sodium nitrate , 16% [w/v] PEG 8000 , 4% [v/v] glycerol , 0 . 1 M Bicine pH 9 ) and equilibrated over reservoir solution . The complex crystallizes in space group P6122 , with unit cell dimensions a , b = 77 . 6 Å , c = 356 . 0 Å , Vm = 2 . 59 Å3 Da–1 , and one molecule of IscS and one molecule of IscU in the asymmetric unit . For data collection , crystals were transferred to reservoir solution supplemented with 15% ( v/v ) ethylene glycol and flash cooled in a nitrogen stream at 100 K ( Oxford Cryosystems , Oxford , UK ) . Diffraction data for both crystal forms of IscS-TusA were collected at the sector 31-ID beamline ( LRL-CAT ) , Advanced Photon Source , Argonne National Laboratory . Data for the IscS-IscU crystal were collected at the CMCF 08ID beamline , Canadian Light Source , Saskatoon , Saskatchewan . Data integration and scaling were performed with HKL2000 [57] . The structures were solved by molecular replacement with the program Phaser [58] using the previously-reported E . coli IscS ( PDB code 1P3W ) and TusA ( PDB code 1DCJ ) structures as the search models . Refinement was carried out with the programs Refmac5 [59] and Phenix [60] , and the models were improved by interspersed cycles of fitting with Coot [61] . The structures were refined applying group B-factors ( one per chain for low resolution and one per residue for medium resolution ) . The translation-libration-screw ( TLS ) model was applied near the end of refinement . For IscS-TusA form 1 the final R-work is 0 . 222 and R-free is 0 . 240 at 2 . 45 Å resolution . The residues 327–332 and 391–404 in IscS subunit A , 329–332 and 393–404 in subunit B , and residues 1–3 and residue 81 in both TusA molecules are disordered and were not modeled . For crystal form 2 the R-work is 0 . 207 and R-free is 0 . 249 at 2 . 45 Å resolution . The residues 329–332 and 393–404 in IscS and 1–3 and 80–81 of TusA are disordered and were not modeled . The IscS-PLP structure was refined at 2 . 05 Å resolution to R-work of 0 . 198 and R-free of 0 . 239 . The disordered region included residues 328–332 and 399–404 in chain A and 328–332 and 394–404 in chain B . In all IscS molecules the loop 322–333 , carrying the essential catalytic Cys328 that accepts the S atom in the persulfated form , extends away from the body of IscS and is less well ordered . The structure of the IscS-IscU complex was also solved by molecular replacement with the same search model for IscS and using the IscU search model ( PDB code 2Z7E ) with program Phaser and was refined using tight geometric restraints at 3 . 0 Å resolution to R-work of 0 . 225 and R-free of 0 . 269 . The residues 328–332 and 394–404 in IscS and residues 1 , 127–128 in IscU were not modeled . In each structure the tips of several sidechains , mostly lysines , arginines , and glutamates , were also disordered and were not included in the models . All models have good stereochemistry ( Table 1 ) as analyzed with PROCHECK [62] . Coordinates have been deposited in the RCSB Protein Data Bank with accession codes 3LVJ for IscS-TusA form 1 , 3LVK for IscS-TusA form 2 , 3LVL for IscS-IscU , and 3LVM for IscS structures , respectively . Data collection and refinement statistics are summarized in Table 1 . The SAXS measurements were carried out using an Anton Paar SAXSess camera equipped with a PANalytical PW3830 X-ray generator and a Princeton CCD detector . The beam length was set to 18 mm and the beam profile was recorded using an image plate for subsequent desmearing . Data for the IscS-IscU complex were collected at 4°C with protein concentrations of 4 . 5 mg/ml ( 10 h ) , 10 mg/ml ( 2 h ) , and 21 mg/ml ( 2 h ) . For the IscS-TusA complex , a data set was recorded at 4°C for 30 min at 22 mg/ml . Dark current correction , scaling , buffer subtraction , and desmearing were performed using the Anton Paar software SAXSquant 3 . 0 . Data sets recorded at different concentrations for IscS-IscU were merged in Primus after removal of the lowest resolution shell ( 0 . 012–0 . 12 Å−1 ) for the 10 and 21 mg/ml data sets , for which Guinier plots showed larger Rg values ( ∼39 Å ) indicating concentration-dependent oligomerization . The data sets were binned ( 5∶1 ) in the range of 0 . 012–0 . 35 Å−1 and fitted directly against predicted scattering calculated from atomic coordinates using the program CRYSOL ( http://www . embl-hamburg . de/ExternalInfo/Research/Sax/crysol . html ) . Experimental Rg values were estimated from Guinier plots , while calculated Rg values were determined using CRYSOL . Oligonucleotide primers were designed according to the QuikChange site-directed mutagenesis method ( Stratagene ) and synthesized by Integrated DNA Technologies . Using the plasmids carrying the wild-type genes as templates , the mutagenesis was performed according to the manufacture's instructions . E . coli DH5α was transformed with the mutagenized plasmids . Plasmids were isolated from the transformants and verified by DNA sequencing . E . coli BL21 ( DE3 ) were then transformed with plasmids containing the confirmed point mutations for protein expression . Mutants of IscS and all binding partners were expressed following the same protocol used for the wild-type counterparts . To follow the interactions between IscS and its partners , we used His6-IscS and partner proteins fused to an N-terminal , TEV-cleavable GST tag . For a specific protein pair , cell pellets from 250 ml individual cultures were mixed , sonicated , centrifuged , and the protein supernatant loaded onto a 250 µl glutathione Sepharose column . Beads were washed with 3 column volumes of buffer ( 50 mM Tris-HCl pH 8 , 200 mM NaCl , 2% ( v/v ) glycerol , except for CyaY where 50 mM NaCl was used ) . For the IscS-IscU pair , the GST-tag on IscU was cleaved prior to elution in order to distinguish its molecular weight from that of IscS . As a positive control , in each case co-purification of the wild-type protein complex was performed in parallel . Proteins retained on the beads or in the case of IscS-IscU , the eluted protein sample , were analyzed by SDS-PAGE . The tusA and iscS genes were deleted by targeted homologous recombination [63] using the oligonucleotide primers TusA ( F ) , TusA ( R ) , IscS ( F ) , and IscS ( R ) ( Table S1 ) . The BW25113 [63] derivative strains were named IC6085 ( BW25113 tusA::kan ) and IC6087 ( BW25113 iscS::kan ) . tRNA from the wild-type and mutant strains carrying pMJ623 , pMJ683 , or their derivative plasmids was purified and degraded to nucleosides as previously described [64] . The hydrolysate was analyzed by HPLC [65] using a Develosil C30 column ( 250×4 . 6 mm; Phenomenex Ltd ) . Western blot analysis to detect GST-TusA , GST-CyaY , GST-IscX , and GroEL proteins was performed with anti-GST ( a generous gift from R . Pulido ) and anti-GroEL antibodies ( Calbiochem ) . | Sulfur is incorporated into the backbone of almost all proteins in the form of the amino acids cysteine and methionine . In some proteins , sulfur is also present as iron–sulfur clusters , sulfur-containing vitamins , and cofactors . What's more , sulfur is important in the structure of tRNAs , which are crucial for translation of the genetic code from messenger RNA for protein synthesis . The biosynthetic pathways for assembly of these sulfur-containing molecules are generally well known , but the molecular details of how sulfur is delivered from protein to protein are less well understood . In bacteria , one of three pathways for sulfur delivery is the isc ( iron-sulfur clusters ) system . First , an enzyme called IscS extracts sulfur atoms from cysteine . This versatile enzyme can then interact with several proteins to deliver sulfur to various pathways that make iron–sulfur clusters or transfer sulfur to cofactors and tRNAs . This study describes in atomic detail precisely how IscS binds in a specific and yet distinct way to two different proteins: IscU ( a scaffold protein for iron–sulfur cluster formation ) and TusA ( which delivers sulfur for tRNA modification ) . Furthermore , by introducing mutations into IscS , we have identified the region on the surface of this protein that is involved in binding its target proteins . These findings provide a molecular view of the protein–protein interactions involved in sulfur transfer and advance our understanding of how sulfur is delivered from one protein to another during biosynthesis of iron–sulfur clusters . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology/microbial",
"growth",
"and",
"development",
"microbiology/microbial",
"growth",
"and",
"development",
"biochemistry/macromolecular",
"assemblies",
"and",
"machines"
] | 2010 | Structural Basis for Fe–S Cluster Assembly and tRNA Thiolation Mediated by IscS Protein–Protein Interactions |
Tomato yellow leaf curl virus ( TYLCV ) is a devastating disease of tomato ( Solanum lycopersicum ) that can be effectively controlled by the deployment of resistant cultivars . The TYLCV-resistant line TY172 carries a major recessive locus for TYLCV resistance , designated ty-5 , on chromosome 4 . In this study , the association between 27 polymorphic DNA markers , spanning the ty-5 locus , and the resistance characteristics of individual plants inoculated with TYLCV in 51 segregating recombinant populations were analyzed . These analyses localized ty-5 into a 425 bp region containing two transversions: one in the first exon of a gene encoding the tomato homolog of the messenger RNA surveillance factor Pelota ( Pelo ) , and a second in its proximal promoter . Analyses of susceptible and resistant lines revealed that the relative transcript level of the gene remained unchanged , regardless of whether the plants were infected with TYLCV or not . This suggests that the polymorphism discovered in the coding region of the gene controls the resistance . Silencing of Pelo in a susceptible line rendered the transgenic plants highly resistant , while in the resistant line TY172 had no effect on symptom development . In addition , over-expression of the susceptible allele of the gene in the resistant TY172 line rendered it susceptible , while over-expression of the resistant allele in susceptible plants had no effect . These results confirm that Pelo is the gene controlling resistance at the ty-5 locus . Pelo , implicated in the ribosome recycling-phase of protein synthesis , offers an alternative route to promote resistance to TYLCV and other viruses .
Tomato yellow leaf curl virus ( TYLCV ) is one of the most devastating viruses of cultivated tomatoes , Solanum ( S . ) lycopersicum . Although first identified in the eastern Mediterranean [1] , it has spread into almost all tropical and subtropical regions [2 , 3] . TYLCV induces severe cupping of apical leaves , yellowing , and stunting , resulting in considerable yield losses . TYLCV has become a major limiting factor to tomato production in major tomato-growing areas , including: China , Mexico , Florida and California [4 , 5] . TYLCV is a monopartite begomovirus ( family Geminiviridae ) transmitted by the whitefly Bemisia tabaci ( Gennadius ) . Population outbreaks of whiteflies are often associated with a high incidence of the disease [6] . The virus genome is composed of a single circular single-stranded DNA molecule of about 2 , 800 nucleotides . Management of TYLCV is difficult because its whitefly vector populations can reach enormous numbers . Therefore , breeding TYLCV-resistant tomato cultivars provides an attractive , environmentally sound , strategy to reduce yield losses inflicted by the virus [4 , 7–9] . Considerable efforts have been invested in breeding TYLCV-resistant tomato cultivars [4] . As all cultivated tomato accessions are susceptible to the disease , wild tomato species were screened to identify , map and introgress resistance loci into S . lycopersicum . Among these are: Ty-2 that was introgressed from S . habrochaites , ty-5 presumably from S . peruvianum , and Ty-1 , Ty-3 and Ty-4 from S . chilense accessions [10–17] . Because Ty-1 and Ty-3 are allelic [18 , 19] , the number of available genes conferring TYLCV resistance is quite limited . Recently , the gene responsible for TYLCV-resistance at the Ty-1/Ty-3 locus was identified and shown to code for an RNA-dependent RNA polymerase ( RDR ) [19] . Moreover , it was shown that Ty-1/Ty-3 confers resistance to TYLCV by increasing cytosine methylation of the viral genome , indicating that the resistance conferred by this locus acts through viral transcriptional gene silencing [20] . Thus far , no other gene conferring TYLCV resistance has been unambiguously identified . Line TY172 , carrying ty-5 , is thought to be derived from four different wild tomato accessions , three of S . peruvianum: PI 126926 , PI 126930 , PI 390681 , and one of S . Arcanum: LA0441 [21 , 22] . The breeding procedure yielding TY172 was described before [17 , 21] . TY172 is highly resistant to TYLCV: it shows no disease symptoms following infection and contains low levels of viral DNA [23] . TY172 exhibited the highest level of resistance in a field trial that compared fruit yield of various resistant accessions following TYLCV-inoculation [23] . It was also found that TY172 , probably due to its high level of TYLCV-resistance , is a poor source for viral acquisition and transmission by whiteflies [24] . These characteristics emphasize the high potential of utilizing TY172 in breeding TYLCV-resistant tomato cultivars . Classical segregation studies suggested that resistance in TY172 is controlled by three genes exerting a partially-dominant effect [21] . However , a study designed to map genes controlling TYLCV-resistance in TY172 showed that resistance is conferred by a previously unknown major recessive quantitative trait locus ( QTL ) , termed ty-5 , that maps to chromosome 4 and four minor OTLs [17] . Recently , a recessive resistance carried by the old commercial cultivar Tyking ( Royal Sluis , The Netherlands ) has been shown to co-localize with ty-5 [25] . The authors suggested that because one of the populations used by Anbinder et al . [17] also displayed a recessive gene action , the resistance in Tyking most likely corresponds to the resistance in TY172 . TYLCV-resistance inherited by TY172 at the ty-5 locus is highly associated with a gene encoding a NAC DOMAIN 1 protein ( Nac1 ) [17] . Nac1 was previously implicated in the replication of another tomato-infecting begomovirus ToLCV , by interacting with its replication enhancer protein ( REn ) in cultivated tomato [26] . It was further shown that ToLCV induce Nac1 expression in infected susceptible cells , and that this up-regulation requires REn . Also , in a transient ToLCV replication system , over-expression of Nac1 resulted in a substantial increase in viral DNA accumulation . These results suggest that Nac1 plays an important role in replication-enhancement of ToLCV and possibly other begomoviruses , including TYLCV , in susceptible plants . Therefore , this gene , or more precisely its homolog in TY172 , was initially referred to as a candidate gene reducing TYLCV replication and thus conferring resistance at the ty-5 locus [17] . The objective of this study was to fine-tune map ty-5 . For this purpose , the associations between 27 polymorphic DNA markers spanning the ty-5 locus , including Nac1 , and the resistance characteristics of individual plants inoculated with TYLCV in segregating populations were analyzed . These analyses , coupled with transgenic confirmation , identified the gene controlling resistance at the ty-5 locus as the tomato homolog of the messenger RNA surveillance factor Pelo .
To fine-tune map the introgression in the TYLCV-resistant line TY172 , carrying the ty-5 gene , we have sequenced 800-to-900 bp fragments of its genome spanning the Nac1 gene region ( first , approximately every 50 Kilo bp ( Kbp ) , then every 10 Kbp and finally every 3 Kbp and at times even in smaller intervals ) . These sequencing results were compared to the sequence of the reference genome Heinz 1706 , known to be susceptible to TYLCV ( build SL2 . 40 in http://solgenomics . net/ ) in order to identify nucleotide polymorphisms . These polymorphisms were validated by sequencing M-82 as well . From these sequence comparisons a core set of 27 markers were generated and used throughout this study ( S1 Table ) . Genomic DNA sequence comparisons between the resistant TY172 line and its susceptible counterparts did not yield significant insertions or deletions usually characterizing DNA sequences of species distantly related to the cultivated tomato , including S . peruvianum . In addition , no polymorphism was detected upstream of the Nac1 promoter gene-sequence beyond those displayed ( S1 Fig ) . These results indicate that the control of TYLCV resistance is exerted either by Nac1 , characterized by Tyrosine212-to-Cysteine substitution in TY172 ( S2 Fig ) , or by a gene located downstream from its position on chromosome 4 . To carry out the map-based analysis of ty-5 , a series of 51 recombinant lines were generated . As this analysis progressed , it became evident that Nac1 is not the gene controlling TYLCV-resistance at the ty-5 locus . For example , a BC1F3 population fixed for the Nac1 allele originating from the susceptible line M-82 and segregating for markers 31 through 0 . 5 still displayed a strong association between the resistance and segregating markers at this range ( Fig 1A ) . On the other hand , a BC2F3 population segregating solely for Nac1 did not display a significant association between the gene and disease symptoms ( Fig 1B ) . These results delimit the ty-5 gene into approximately 351 Kb between Nac1 and the 5 . 8 marker ( S1 Table ) . We have analyzed 5 , 662 plants from different segregating populations following a cross between the resistant TY172 and the susceptible M-82 line . 51 plants displaying recombination events among the polymorphic DNA markers were identified . These plants were allowed to self-pollinate to produce segregating populations that were again genotyped , inoculated and assayed for resistance ( S3 Fig ) . These 51 recombinant plants enabled us to narrow down the resistant locus into a 23 , 250 bp fragment ( in TY172 ) , including the amplicons produced by markers 4 . 4 and 4 . 8 , containing two genes: a Calcium dependent protein kinase ( Cdpk2 ) [27] and the Pelo gene homolog [28] ( S3 Fig ) . The identification of this region can be exemplified as follows: The region between markers 4 . 4 and 4 . 8 was sequenced in TY172 and M-82 and compared to the sequence of Heinz 1706 . Seven single nucleotide polymorphisms ( SNPs ) were found in this region . Further sequencing of the polymorphic region between Pelo and marker 4 . 4 in selected recombinant plants of populations such as the one displayed in Fig 1D have shown that the three SNPs found in this region ( between Pelo and marker 4 . 4 , the region in which Cdpk2 is located ) are unrelated to the resistance . This clearly demonstrated that Cdpk2 is not involved in the resistance , and enabled us to delimit the introgression into a polymorphic 2 , 386 bp region between Pelo and marker 4 . 8 . The SNPs in this region were: a G1-to-A transition , single nucleotide insertion ( A1565 ) , and a T1961-to-A1962 transversion in the promoter region of the Pelo gene in TY172 ( S4 Fig ) . In addition , a T47-to-G transversion in the first exon of Pelo was also observed ( T2385-to-G2386 in S4 Fig ) . Additional sequencing of the polymorphic region between Pelo and marker 4 . 8 in selected recombinant plants obtained from populations such as the one displayed in Fig 1C enabled us to finally delimit the ty-5 gene into a 425 bp region containing two transversions . One transversion is the T1961-to-A1962 in the Pelo proximal promoter region and the other is the T47-to-G in the first exon of the gene ( S4 Fig ) , resulting in a Valine16-to-Glycine substitution in TY172 ( Fig 2 ) . These results show that Pelo is the gene controlling TYLCV resistance at the ty-5 locus and that this resistance can be either attributed to its proximal promoter or its coding region . Noteworthy , other polymorphisms , but none of these SNPs , were found following sequencing of this region in DNA extracted from two-to-three plants each of PI 126926 , PI 126930 , PI 390681 and LA0441 , the four accessions claimed to be the origin of ty-5 ( demonstrated by their Pelo coding sequence in GenBank accession numbers KC567248 through KC567257 ) . Moreover , screening of the 360 tomato genomes provided by Lin et al 2014 [29] showed no polymorphism at the two transversions mentioned above ( the T1961-to-A1962 in the Pelo proximal promoter and the T47-to-G in the first exon of the gene ) . Sequence analysis of a polymorphic region tightly linked to Pelo displayed complete identity between two plants of the old tomato hybrid Tyking genomic DNA with TY172 DNA ( S4 Fig ) . Moreover , the amino-acids sequence of Pelo from Tyking was identical to its counterpart from TY172 ( Fig 2 ) . It is therefore highly likely that ty-5 in TY172 and Tyking originated from the same unknown source . Because to the best of our knowledge Tyking preceded TY172 , we cannot exclude the possibility that ty-5 in TY172 was introgressed from Tyking . Nonetheless , the origin of ty-5 locus in Tyking is yet to be established . To test whether the SNP in the Pelo promoter-region is involved in the resistant phenotype , relative transcript level of the gene was compared between resistant and susceptible genotypes . The results show that relative transcript level of Pelo in the resistant line TY172 was not statistically different from the susceptible line M-82 , whether the plants were infected with TYLCV or not ( Table 1A ) . The relative transcript level of Pelo was also analyzed in another susceptible line used in this study , R13 . The relative transcript level of Pelo in the resistant line TY172 was not statistically different from this susceptible line as well ( Table 1B ) . These results suggest that the SNP in the Pelo promoter region is not associated with TYLCV resistance . To confirm that indeed the SNP in the Pelo coding-region is responsible for the resistance phenotype , transgenic plants over-expressing each of the two Pelo alleles were developed . Two independent populations of transgenic TY172 plants segregating for pBINPelo-M-82 were analyzed ( Table 2 ) . Following inoculation with TYLCV , all transgenic plants displayed disease symptoms while their respective azygous controls were symptomless , similarly to inoculated non-transgenic TY172 plants . The transgenic plants showed an average disease severity index ( DSI ) values of 2 . 2 and 2 . 3 for TYT-10 and TYT-94 , respectively , correlated with an approximately 3 . 5-fold higher average virus copy number ( VCN ) in transgenic plants as compared to their non-transgenic azygous controls ( Table 2 ) . The two transgenic TY172 lines were also assayed for transcript-expression levels of Pelo ( Table 2 ) . In both transgenic lines there was a significant increase in Pelo transcript level as compared to their non-transgenic azygous controls , a five-fold increase in the transgenic line TYT-10 ( a T1 generation ) , and a 91-fold increase in the transgenic line TYT-94 ( a T3 generation ) . These results clearly demonstrate that TY172 plants over-expressing the susceptible allele of Pelo are no longer resistant to TYLCV , although their DSI is lower than that expected from fully susceptible plants . A possible explanation can be the presence of the minor OTLs identified in TY172 [17] . Three independent T1 populations of R13 plants segregating for pBINPelo-TY172 were also analyzed . While RNA level of the transgene was significantly increased in these lines ( 179 , 229 and 173-fold in TYT-002 , TYT-066 and TYT-103 , respectively ) , no statistical difference was found in average DSI values between the transgenic and their azygous counterparts ( Table 3 ) . In all the three lines there was also no statistical difference in VCN between the transgenic plants and their azygous controls . These results show that over-expression of the resistant allele of Pelo does not affect the TYLCV-susceptible R13 plants , which is in agreement with the recessive nature of the ty-5 locus in segregating populations ( Fig 1A , 1C and 1D ) . To further confirm that indeed the Pelo coding-region is responsible for the resistance phenotype , transgenic TYLCV-susceptible ( R13 ) and resistant ( TY172 ) plants harboring a Pelo silencing construct ( pHannibal-Pelo ) were developed . Three independent T1 populations of transgenic R13 plants segregating for the silencing construct pHannibal-Pelo were analyzed ( Table 4 ) . Following inoculation with TYLCV , all the transgenic plants showed no disease symptoms while their respective azygous controls showed severe disease symptoms ( yellowing and cupping of the leaves , most clearly shown in the plant apex ) ( S5 Fig ) , similarly to inoculated R13 control plants ( S6 Fig ) . These transgenic plants showed an average DSI values of practically 0 , correlated with a 2 . 2-to-3 . 5 fold reduction in Pelo transcript level , and a 20-to-60 fold decrease in average VCN in transgenic plants as compared to their non-transgenic azygous controls ( Table 4 ) . However , when the two independent T1 populations of transgenic TY172 plants segregating for the silencing construct were analyzed there were no differences in disease severity between the transgenic plants and their respective azygous controls ( Table 4 and S6 Fig ) . Both transgenic and azygous control alike showed no disease symptoms , with a DSI of 0 , coupled with a six-to-nine fold reduction in Pelo transcript level in the transgenic plants as compared to their non-transgenic azygous controls ( Table 4 ) . Interestingly , despite the lack of difference in DSI , there was a significant difference in virus copy number: the transgenic plants showed a nine-to-tenfold decrease in VCN compared to their non-transgenic azygous control plants ( Table 4 ) . A nearly isogenic BC4F3 segregating-population was developed by crossing TY172 ( RR ) as a maternal line and M-82 ( SS ) as a recurrent paternal line , using Pelo as the sole marker . Non-inoculated homozygous RR plants of this population displayed a significant reduction in fruit size , and a small insignificant reduction in total fruit yield and harvest index compared to the recurrent SS parent ( Table 5 ) . However , when compared to their homozygous SS counterparts the homozygous RR plants displayed a significant reduction of 23% in total yield and a 27% reduction in fruit size , suggesting that the presence of ty-5 may exerts a penalty .
TY172 is a tomato line expressing high level TYLCV-resistance . Based on a classical segregation study , it was suggested that the resistance is controlled by three genes that originated from S . peruvianum [21] . Subsequently , molecular mapping studies showed that TYLCV-resistance in TY172 is controlled by a major recessive QTL , termed ty-5 , and four additional minor QTLs . The major QTL maps to chromosome 4 while the minor QTLs were mapped to chromosomes 1 , 7 , 9 and 11 [17] . This study was designed to identify the gene controlling TYLCV-resistance at the ty-5 locus . To accomplish this , a core set of 27 polymorphic DNA markers were designed . The use of 51 informative recombinant populations enabled us to exclude Nac1 , a candidate gene suggested earlier [17] , and delimits the ty-5 locus into a single gene encoding the tomato homolog of the messenger RNA surveillance factor Pelo . Two SNPs were identified: one in Pelo’s proximal promoter region and the other in its coding sequence . Our results further show that the relative transcript level of Pelo in the resistant TY172 plants was not statistically different from susceptible plants , either inoculated or not . It can therefore be suggested that the control of TYLCV-resistance is exerted by the SNP observed in the Pelo coding sequence , and not by the one observed in its proximal promoter region . These two SNPs and others observed upstream to their location could not be traced in sequences of two-to-three plants of each of the four S . peruvianum accessions claimed to be the origin of ty-5 [21] . This suggests that the Pelo allele residing in TY172 originated from a different source or that it represents a rare allele segregating in one or more of these four accessions . Noteworthy , Pelo and its upstream promoter sequence from the cultivar Tyking , carrying a recessive resistance co-localized with ty-5 [25] , were found identical to TY172 . Although we cannot exclude the possibility that the resistant Pelo allele originated from Tyking , we were not able to trace its origin in wild and other cultivated tomato accessions , including in those of the 360 tomato genomes provided by Lin et al 2014 [29] . Over-expression of the cultivated Pelo allele in TY172 resulted in increased viral titer and disease symptoms , while over-expression of its resistant counterpart in susceptible plants had no effect on TYLCV titer and disease severity . In addition , silencing of Pelo in susceptible plants rendered these transgenic plants highly resistant to TYLCV—the inoculated plants showed practically no disease symptoms , and a 20-to-60 fold reduction in virus titer compared to non-transgenic control plants . Finally , silencing Pelo in resistant TY172 plants had no effect on disease severity . Nonetheless , these transgenic plants had a nine-to-tenfold decrease in virus titer compared to their non-transgenic control plants . Taken together , these results confirm that Pelo controls the TYLCV-resistance at the ty-5 locus . Pelo was recently implicated in the recycling phase of protein biosynthesis as part of eukaryotic and archaeal ribosome recycling complexes containing also an ABC-type ATPase ( Abce1 ) [28] . Ribosome-driven protein biosynthesis has four phases: initiation , elongation , termination and recycling . Interestingly , recessive genes for resistance to plant viruses have been linked to components of the eukaryotic translation initiation complex . Translation initiation factors , particularly the eIF4E and eIF4G protein families , were found to be essential for RNA virus infections [30] . Here we show that a protein implicated in the latest phase of ribosome-driven protein biosynthesis controls the recessive resistance to TYLCV , a DNA virus . Recessive resistance genes are more prevalent controlling resistance to viruses than resistance to fungal or bacterial pathogens which is primarily a dominant genetic trait [30] . Because typical plant viruses encode relatively few proteins , between four-to-ten , and need to recruit many different host components to complete their infection cycle , it was proposed that resistance genes would correspond to mutation or loss of host components required for a stage of the virus life cycle [31] . This fits well with the recessive nature attributed in this study to the Pelo allele variant residing in TY172 . Although many host factors are required for plant virus infections [32] , analysis of recessive resistance identified in the natural diversity of crops has thus far only revealed a group of proteins linked to the translation machinery [30] . TYLCV-resistance controlled by the ty-5 gene Pelo is not an exception , but points to the ribosome recycling phase of protein synthesis , rather than to its initiation , as an intervening step promoting resistance . Recycling of ribosomes for a new round of translation initiation is an essential part of protein synthesis . As recently summarized [28] , Abce1 , mentioned above , can dissociate ribosomes into subunits either after canonical termination by release factors ( Rfs ) , or after recognition of stalled ribosomes by messenger RNA surveillance factors such as Pelo , an eukaryotic Rf1 ( eRf1 ) paralog . Notably , Abce1 is able to physically interact with eRf1 and directly influence its function in stop-codon recognition and peptidyl-transfer RNA hydrolysis . Failure to accomplish or fully complete this task , anticipated in a recessive mutant , would most probably negatively affect viral as well as host-plant protein synthesis . This in turn may result in slower infection progression , but may also negatively impact aspects of host plant development or its horticultural performances [33] . A screen-house study , carried out in this study , demonstrate that uninfected nearly-isogenic BC4F3 plants , homozygous for the Pelo allele derived from TY172 , displayed a significant reduction in fruit size , and a small but not significant reduction in total fruit yield and harvest index compared to the recurrent susceptible parent ( Table 5 ) . This suggests that the presence of ty-5 may exert a small penalty on yield . Whether these negative effects remain in more advanced BC generations is yet to be elucidated . Noteworthy , this discussion is limited to non-inoculated plants , to the determinate growth-habit characterizing TY172 and M-82 plants , and to a single genetic background ( M-82 ) , while excluding possible contribution that hybrid vigor may have towards increased yield in hybrid commercial plants carrying ty-5 . It should be clarified , that under TYLCV-inoculation , ty-5 has a clear advantage over susceptible plants [17 , 23] . In a recent forward genetic screen for Drosophila mutants that are resistant to Drosophila C virus ( DCV ) , a virus resistant line was found which had a deficient expression of the Pelo gene [34] . The Pelo deficiency was found to limit the high level synthesis of the DCV capsid protein . It was suggested that the reduction in synthesis of viral capsid is due to the role of Pelo in dissociation of stalled 80S ribosomes and clearance of aberrant viral RNA and proteins [34] . It should be noted however , that DCV is an RNA virus , completely different from TYLCV , which is a ssDNA virus . To the best of our knowledge , Pelo and its protein product have never been implicated in virus resistance in plants and thus offer an alternative route to obtain such resistance . Our suggestion that its effect may be expressed through its role in protein translation machinery is promising . However , a direct interaction of Pelo with proteins involved in viral replication cannot be excluded .
5 , 662 segregating F2 , F3 , F4 , BC1F1 , BC1F2 , BC1F3 , BC1F4 , BC1F5 , BC2F2 , BC2F3 and BC2F4 seedlings originating from a cross between TY172 and M-82 ( LA3475 ) , were inoculated with TYLCV and genotyped with polymorphic markers spanning the ty-5 locus . M-82 was chosen as the recurrent parent due to its determinate stature , facilitating working with large number of plants . The inoculation experiments were carried out in 8x16 sowing trays . Eight seedlings of the two parental lines were included in each tray while their F1 hybrid plants were included sporadically . Twenty-one days post inoculation ( DPI ) , the seedlings were transplanted to the field or to 50-mesh screen-houses . Parental lines and their F1 hybrids were planted in a randomized block design , five or more plants in three blocks , while their segregating counterparts were planted at random . In the field plants were grown in rows , one m apart , allowing 50 cm space between plants , while in screen-houses in eight L pots , similarly arranged . Plants were grown using standard local cultural practices , including drip irrigation and fertilization . The effect of the Pelo allele derived from TY172 on yield of non-inoculated segregating BC4F3 plants was estimated in a randomized-block screen-house experiment ( three blocks , five plants per genotype per block ) . These BC4F3 plants , developed with M-82 as a recurrent parent and Pelo as the sole marker , were compared to M-82 . Total fruit yield , fruit number , average fruit weight , plant weight and harvest index ( fruit-yield/plant-weight ratio ) were recorded for each plant . Other lines and accessions used: 1 . Heinz 1706—a TYLCV-susceptible determinate line obtained from the Tomato Genetics Resource Center ( TGRC at http://tgrc . ucdavis . edu/ ) . This line was initially used to sequence the tomato genome ( SGN , http://solgenomics . net/ ) ; 2 . R13—a TYLCV-susceptible indeterminate line ( Hazera Genetics , Berurim , Israel ) ; 3 . LA1589 , LA0441 , PI 126926 , PI 126930 , and PI 390681—LA accessions were obtained from TGRC while PI accessions were obtained from the U . S . Department of Agriculture , Plant Genetic Resources Unit , Geneva , NY; 4 . Seeds of the TYLCV-resistant hybrid Tyking were generously provided by Jaap Hoogstraten ( Monsanto ) . Whitefly colonies ( Bemisia tabaci , biotype B ) , reared on cotton plants ( Gossypium hirsutum ) , were used for TYLCV ( Genbank accession No . X15656 ) inoculation as described [23 , 35] . Thereafter , plants were sprayed with imidacloprid ( Bayer , Leverkusen , Germany ) and held in an insect-proof greenhouse at 26–32°C . TYLCV-induced symptoms were evaluated for each plant separately at 28 and 42 DPI according to a disease severity index ( DSI ) of 0-to-4 as follows: ( 0 ) no visible symptoms , inoculated plants show same growth and development as non-inoculated plants; ( 1 ) very slight yellowing of leaflet margins on apical leaf; ( 2 ) some yellowing and curling of leaflet ends; ( 3 ) a wide range of leaf yellowing , curling and cupping , with reduction in size , yet plants continue to develop and ( 4 ) very severe plant stunting and yellowing , pronounced cupping and curling , plants stop growth [21 , 24] . Genomic DNA was extracted from individual plants according to [36] . A 4394 base-pair ( bp ) genomic region spanning the Nac1 gene sequence in M-82 was blasted against the tomato sequence database at the Solanaceae Genomics Network ( SGN , http://solgenomics . net/ ) and found fully homologous to the 8 , 734 , 372–8 , 738 , 765 bp region of the scaffold DNA sequence file SL2 . 40sc03604 , and 2 , 854 , 539–2 , 858 , 932 bp region of the tomato chromosome 4 sequence in the DNA sequence file SL2 . 40ch04 ( version SL2 . 40 ) . Sequences in SL2 . 40sc03604 and SL2 . 40ch04 were used to design polymerase chain reaction ( PCR ) primers expected to amplify 800-to-900 bp fragments of the tomato genome spanning the Nac1 gene region: first , approximately every 50 Kilo bp ( Kbp ) , then every 10 Kbp and finally every 3 Kbp and at times less . Altogether , 257 such fragments were sequenced . Sequence analysis and locus-specific primer design were carried out with the DNAMAN sequence analysis software v4 . 1 ( Lynnon BioSoft , Québec , Canada ) . DNA primers were purchased from Molecular Biology Center ( Ness-Ziyyona , Israel ) . The primers designed were used to PCR-amplify genomic DNA of TY172 and M-82 plants; amplification products were visualized by electrophoresis , extracted from the gel and directly sequenced according to [17] . These sequences were compared to the sequence of Heinz 1706 ( http://solgenomics . net/ ) in order to detect single nucleotide polymorphisms ( SNPs ) . These polymorphisms were further used to design polymorphic DNA markers , utilizing restriction endonucleases when necessary , that were analyzed for association with the resistance trait in segregating populations . Altogether , a core set of 27 polymorphic DNA markers , spanning the ty-5 locus , were utilized by PCR ( S1 Table ) . Following amplification , PCR products were digested and visualized by electrophoresis according to [17] . We used a Melting curve SNP genotyping method ( McSNP ) with primers designated as McSNP F and R in S1 Table . Primers design was carried out by DYN R&D ( Qesariyya , Israel ) . The McSNP genotyping reaction , described previously [37] , was calibrated and carried out by DYN R&D using the LightCycler 480 instrument ( Hoffmann-La Roche , Basel , Switzerland ) . Initial reaction conditions were: incubation at 95°C 10 min , followed by 50 cycles of 95°C 10 sec , a touch-down annealing 63°C→56°C 10 sec ( -1 . 5°C per cycle ) and 72°C 10 sec . Melting curve reaction conditions were: 95°C 1 min , 40°C 2 min , and 40–75°C degree , five acquisitions per degree . Results were analyzed using the LightCycler 480 SW 1 . 5 software ( Hoffmann-La Roche ) , by DYN R&D utilizing its melt-curve analysis . Construction of Pelo silencing vector: to silence Pelo , a pHannibal vector [38] expressing a sense and anti-sense fragment of the gene was constructed in two steps . In the first step , a 576 bp fragment of the Pelo gene ( cDNA coordinates 560 to 1136 ) was amplified by PCR using the forward primer SPeloF-XhoI ( 5'-AGACTCGAGGACAATGTTCTACAGGCCTTTG-3' ) , containing a XhoI restriction site , and the reverse primer SPeloR-KpnI ( 5'-GACGGTACCCATCTCAAT GTCTTCCAGCTC-3' ) , containing a KpnI site . This fragment was cloned into the unique XhoI and KpnI sites present in the sense oriented arm of pHannibal . In the second step , the same 576 bp fragment of the gene was amplified by PCR performed with the forward primer SPeloF-XbaI ( 5'-ATCTAGAGACAATGTTCTACAGGCCTTTG-3' ) , containing a XbaI site , and the reverse primer SPeloR-ClaI ( 5'-CATCGATCATCTCAATGTC TTCCAGCTC-3' ) , containing a ClaI site . This fragment was cloned into the unique XbaI and ClaI sites present in the anti-sense oriented arm of pHannibal , thus creating pHannibal-Pelo . To create a binary vector , pHannibal-Pelo was cloned under the cauliflower mosaic virus ( 35S ) promoter and the nitric oxide synthase transcriptional terminator into the NotI site of the pBIN vector . Transformations were carried out on cotyledon cuttings of TY172 and the susceptible R13 lines with Agrobacterium tumefaciens strain EHA105 as previously described [39] . R13 was chosen as the susceptible control due to its ease of transformation . Moreover , sequence analysis of Pelo in R13 revealed that it is identical to M-82 . Pelo cDNA was cloned from both TY172 and M-82 plants , and inserted into a pBIN vector under the control of the 35S promoter . To create a pBIN expression vector , the cassette containing the 35S promoter , omega enhancer , and the NOS terminator was cloned into the HindIII-EcoRI sites of pBINPLUS . To create the plasmids pBINPelo-M-82 and pBINPelo-TY172 , the Pelo gene was PCR amplified using the primers PeloF ( 5’-CTAGGATCCatgaagattgttcgtagag-3’ ) , containing a BamHI restriction endonuclease site , and PeloR ( 5’-CTAGCGGCCGCATCACATCTCAATGTCTTC-3' ) , containing a NotI site . The amplification product was restricted with both BamHI and NotI and cloned into the appropriate sites of the pBIN vector . Transformations were carried out as described above . To validate incorporation of either the Pelo silencing construct or the over-expression constructs , DNA samples extracted from individual transformed plants served as templates in PCRs using a primer complementary to the 35S promoter ( 5’-CCTTCGCAAGACCCTTCCTCT-3’ ) and a primer complementary to 3’ of the Pelo gene sequence ( 5’-CTAGCGGCCGCATCACATCTCAATGTCTTC-3' ) for the silencing construct , while for the two Pelo transgenes a primer complementary to the Pelo gene sequence ( 5’-CTAGCGGCCGCATCACATCTCAATGTCTTC-3' ) was used . The PCRs were performed in a volume of 20 μl containing 15 ng of template DNA , 10 pmol of each primer , 0 . 2 mM of each dNTP , 2 mM MgCl2 , 0 . 5 U of Taq DNA polymerase , and 1XPCR-buffer . The PCRs conditions were: 94°C 3 min , followed by 35 cycles of 94°C 30 s , 60°C 30 s , and 72°C 1 min . Final elongation was at 72°C 10 min . Amplification products were visualized by electrophoresis . TYLCV DNA copy-number in plants was determined using quantitative Real-Time PCR ( qRT-PCR ) . Total DNA was extracted from plant apices . TYLCV primers were designed using the PRIMER 3 procedure ( http://workbench . sdsc . edu/ ) : TYRT2F = 5'-GCTGATCTGCCATCGATTTG-3' and TYRT2R = 5'-GGTTCTTCGACCTGGTATC-3' forming a 147 bp amplicon . The qRT-PCR was carried out on a Corbett Rotor-Gene 6000 ( Qiagen , Düesseldorf , Germany ) with the following profile: 40 cycles of 95°C 10 s , 60°C 15 s , and 72°C 20 s; qRT-PCR reactions ( 12 μl volume ) included 3 μl of plant DNA , 6 μl of SYBR Fast Universal Readymix Kit ( Kapa Biosystems , Boston , MA ) , and 0 . 125 μM of each primer . DNA extracted from non-infected plants and water served as negative controls . Each qRT-PCR reaction was run in duplicate , with 5 replications per treatment . DNA of each sample was extracted from three different infected plants . For standard curve , PCR amplicon was cloned into pGEM-T Easy ( Promega , Madison , WI ) . The plasmids were extracted using Plasmid Miniprep kit ( Qiagen ) and linearized by digesting with PstI . Gel-extracted fragments were quantified and used to create standard curves . Dilution series were performed by copy number following methods recommended by Applied Biosystems ( Foster City , CA ) . Cycle threshold and copy number were determined using Corbett Rotor-Gene 6000 Series software . Amplification was followed by melt-curve analysis . Pelo relative transcript levels were determined by qRT-PCR . Total RNA was extracted from tomato apex using TRI-reagent ( Sigma-Aldrich , St . Louis , MO ) and DNA contaminants were digested with TURBO DNA-free DNAase ( Ambion , Austin , TX ) . The remaining RNA was used as template for cDNA synthesis using the Revertaid first strand cDNA synthesis Kit ( Fermentas ) . Pelo primers were designed using the Primer 3 procedure . Primers used for the qRT-PCR reaction of the Pelo transcript level were: PeloRTF = 5'-CCATGAGCGTCTGGCTATTC-3' and PeloRTR = 5'-GGAGACATGCATTGACGAGA-3' , forming a 150 bp amplicon . The qRT-PCRs ( 12 μl volume ) were performed as outlined above , with the following profile: 40 cycles of 95°C 10 s , 58°C 15 s , and 72°C 20 s . 18S ribosomal RNA was used as reference utilizing the following primers: 18SF = 5'-GCGACGCATCATTCAAATTTC-3' and 18SR = 5'-TCCGGAATCGAACCCTAATTC-3' . qRT-PCR analyses were performed using the Rotor-Gene Q detection system and data was collected and analyzed with the Rotor-Gene 6000 software version 1 . 7 . 28 ( Qiagen ) . Relative abundance of Pelo transcripts were calculated by the formula: 2- ( CT_Pelo-CT_18S ) , where CT represents the fractional cycle number at which the fluorescence crosses a fixed threshold ( usually set on 0 . 1 ) . The association between DNA markers and DSI scores were evaluated by analyses of variance and chi-square ( χ2 ) with the JMP statistical discovery software ( SAS Institute , Cary , NC ) . An excellent agreement was found between χ2 and the respective analyses of variance; therefore , analyses of variance are presented throughout this manuscript . Differences in the relative abundance of the Pelo transcripts between genotypes were analyzed , following transformation to their natural logarithm , by analyses of variance using JMP software . Differences among means were statistically evaluated based on Tukey-Kramer Honestly Significant Difference test [40] . | Tomato is one of the most important food crops worldwide , providing phytonutrients and color to our diet . Tomato yellow leaf curl virus ( TYLCV ) is one of the most devastating viruses of cultivated tomatoes and a key limiting factor to tomato production in major tomato-growing areas . The management of TYLCV is difficult because its whitefly vector populations can reach high numbers and relies heavily on the application of hazardous chemicals ( insecticides ) . Breeding TYLCV-resistant tomato cultivars provides an attractive , environmentally friendly strategy to reduce yield losses caused by the virus . Considerable efforts have been invested in breeding TYLCV-resistant tomato cultivars . The discovery and utilization of genes controlling resistance can expedite the breeding process and highlight innate means to promote such resistance . Here we report the discovery of the gene controlling TYLCV resistance at the ty-5 locus . It was found that the messenger RNA surveillance factor Pelo is responsible for this resistance . Our discovery , together with the recent discovery of the gene controlling resistance at the Ty-1/Ty-3 locus , can lead to precision pyramiding of these two genes to enhance the spectrum of resistance of tomato plants to TYLCV , and possibly to other begomoviruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | A Novel Route Controlling Begomovirus Resistance by the Messenger RNA Surveillance Factor Pelota |
The objective of this study was to assess the validity of the new dengue classification proposed by the World Health Organization ( WHO ) in 2009 and to develop pragmatic guidelines for case triage and management . This retrospective study involved 357 laboratory-confirmed cases of dengue infection diagnosed at King Abdulaziz University Hospital , Jeddah , Saudi Arabia over a 4-year period from 2014 to 2017 . The sensitivity of the new classification for identifying severe cases was limited ( 65% ) but higher than the old one ( 30% ) . It had a higher sensitivity for identifying patients who needed advanced healthcare compared to the old one ( 72% versus 32% , respectively ) . We propose adding decompensation of chronic diseases and thrombocytopenia-related bleeding to the category of severe dengue in the new classification . This modification improves sensitivity from 72% to 98% for identifying patients who need advanced healthcare without altering specificity ( 97% ) . It also improves sensitivity in predicting severe outcomes from 32% to 88% . In conclusion , the new classification had a low sensitivity for identifying patients needing advanced care and for predicting morbidity and mortality . We propose to include decompensation of chronic diseases and thrombocytopenia-related bleeding to the category of severe dengue in the new classification to improve the sensitivity of predicting cases requiring advanced care .
Dengue fever ( DF ) is the most prevalent arthropod-borne viral disease in humans and one of the major re-emerging communicable diseases . The World Health Organization ( WHO ) estimates that around 50 million dengue infections occur annually and approximately 2 . 5 billion of the world's population live in dengue-endemic areas [1] . Dengue virus is endemic in Saudi Arabia primarily in the western and southern provinces [2 , 3] . It is a well-recognized cause of seasonal outbreaks in Jeddah and Makkah [4–7] . Two large epidemics occurred in Jeddah and Makkah; the first in 2011 , when 2569 cases were reported , and the second , in 2013 when 4411 cases including 8 deaths were reported [6] . Dengue was also reported in other regions of Saudi Arabia , including Al-Madinah ( 2009 ) , and Aseer and Jizan ( 2013 ) [2 , 7] . Since the 1970s , dengue has been conventionally classified into four main categories ( Table 1 ) : non-classical DF , classical DF , dengue hemorrhagic fever ( DHF ) , and dengue shock syndrome ( DSS ) . DHF definition requires the presence of four criteria: fever , thrombocytopenia ( <100 , 000 platelets/mm3 ) , hemorrhagic manifestations , and plasma leakage manifesting as accumulation of fluids in the peritoneal , pleural , or pericardial spaces , lower limb edema , hypoalbuminemia , or hemoconcentration . Several studies reported lack of correlation between the categories of the conventional classification and the disease severity [8–10] . Despite high specificity of the DHF category , the sensitivity is unacceptably low in detecting severe cases of dengue that require specialized care and monitoring in a hospital setting [11–13] . As a consequence , a global expert consensus meeting at WHO in 2008 accorded on a new classification of DF . The revised classification divides dengue into two categories ( Table 2 ) : non-severe and severe dengue ( SDF ) ; the non-severe dengue is further divided into two categories: dengue with warning signs ( D+W ) and dengue without warning signs ( D-W ) . The new classification was developed based on the level of clinical severity to establish management guidelines and to facilitate dengue reporting and surveillance . Warning signs were proposed to facilitate triage and early detection of potentially severe cases that need hospitalization , particularly in primary care settings and during outbreaks . Several studies were conducted to assess the utility of the new dengue classification scheme in clinical practice [14–19] . After the release of the new WHO classification in 2009 , the level of care required by patients with dengue was used by researchers as a gold standard to grade the severity of the illness to identify patients with severe disease that is likely to require high level of healthcare as in-patients and those with milder disease that can be managed as out-patients [11 , 14] . Such classification would conceivably improve outcome and reduce cost of healthcare . Additionally , the level of care is the most reliable indicator of severity in retrospective studies that were primarily used to assess the utility of the new classification in clinical practice . However , current data remain insufficient to establish the validity of this classification in predicting or identifying severe dengue cases that may need close observation or hospitalization for proper management . This study aimed to assess the validity of the new dengue classification scheme based on data from Jeddah city as part of global evaluation of the new dengue classification .
This retrospective study included patients with dengue virus infection reported to the infection control unit at King Abdulaziz University Hospital ( KAUH ) , Western Saudi Arabia over a 4-year period from January 2014 through December 2017 . Only laboratory-confirmed cases presenting within 7 days of disease onset were included in the study . The diagnosis was confirmed if at least one of the following criteria was met in acute phase serum: ( 1 ) positive reverse transcription polymerase chain reaction ( RT-PCR ) , ( 2 ) positive serology for dengue IgM , or ( 3 ) positive dengue-specific non-structural antigen-1 ( NS1 ) . Exclusion criteria included patients who presented 7 days after the onset of symptoms , or those who were transferred to other hospitals , or whose data were unavailable . Persistent vomiting was defined as vomiting at least 5 times per day or vomiting everything the patient ingested . Shock was defined as tachycardia ( pulse rate > 100 beats/minute ) with either hypotension-for-age or narrow pulse pressure ( <20 mmHg ) . Severe bleeding was defined as major bleeding ( hematemesis , melena , or menorrhagia ) associated with systolic hypotension , haemoglobin <8 g/dL , or a drop of haemoglobin of >2 g/dL , or bleeding that required blood transfusion . Renal impairment was defined as serum creatinine rise of at least 50% over the baseline that failed to improve after two days of re-hydration . The baseline creatinine was defined as the minimum value following two days of re-hydration . Peak creatinine was defined as the highest creatinine value recorded following two days of re-hydration . The baseline haematocrit ( HCT ) value was defined as the minimum HCT value following a minimum of two days of re-hydration provided that the patient had passed the 6th day of illness . If no baseline HCT could be defined for a given patient , the hospital's average value of normal HCT ( 41% for adults and 42% for pediatrics ) was used as the estimated baseline value . Peak HCT was the maximum HCT value recorded during hospital stay . Patients' clinical outcomes were determined using data from the time of hospital presentation to the time of hospital discharge or demise , and from any subsequent follow up data when available . Warning signs were only considered at patient’s presentation . The day patients developed severe dengue was documented . Lethargy was not included as a warning sign due to inadequate documentation of this symptom in the patients’ records . Therapeutic interventions and healthcare required by patients were classified into three levels: level I , included patients who were treated on outpatient basis; level II , included hospitalized patients who received intravenous fluids for rehydration and/or those who received platelets due to thrombocytopenia that was not associated with major mucosal bleeding; level III , included hospitalized patients who required intravenous fluids for resuscitation , mechanical ventilation , blood transfusion , inotropic support , or specific treatment for organ failure . The authors determined the level of care ( the gold standard ) for each patient based on the maximum healthcare required by the patient any time during his/her entire hospital stay . For instance , if a patient with dengue was stable at presentation but 2 days later the patient required ICU admission , level III of care ( patients requiring intensive care ) will be assigned to this patient . The aim of the classification was to determine the level of care the patient is most likely to require and to predict severe dengue before it happens . The gold standard was assessed during the phase of data collection ( interventions needed were answered as yes or no ) and phase of statistical analysis ( patients were grouped as level I , II , III according to the pre-set definitions in the research proposal ) . Assignment of patients to the level of care they required was reviewed and confirmed by the senior author ( TAM ) who is a professor of medicine and infectious disease with a vast and long experience in managing dengue fever . Data were analyzed using IBM SPSS , version 22 . P values of <0 . 05 were considered statistically significant . Chi-square test was used to compare categorical variables . Diagnostic values including sensitivity , specificity , positive predictive value ( PPV ) , and negative predictive value ( NPV ) of the old and the new WHO classifications for identifying severe cases were analyzed in reference to the level of healthcare the patients required . Thus , the level of healthcare required by patients was the basis for determining severity of dengue in a retrospective manner . For the old classification , patients who were classified as DHF/DSS were considered to have severe dengue , while those classified as non-classical or classical DF were considered to have non-severe dengue . For the new classification , patients classified as SDF were considered to have severe dengue and those classified as dengue with or without warning signs were considered to have non-severe DF . With respect to the gold standard ( management level ) , severe cases were defined as patients who were managed with level III care , whereas non-severe cases were patients who required level I or II care . The study was approved by the Institutional Review Board of King Abdulaziz University Hospital , Jeddah , Saudi Arabia . All data analyzed were anonymized .
Of 471 laboratory-confirmed dengue cases , 357 met the inclusion criteria . Of those 357 cases , 244 ( 68% ) were males; 53 ( 15% ) were children ( <15 years old ) with a mean ± standard deviation ( SD ) age of 8 ± 4·4 years , and 304 ( 85% ) were adults ( ≥15 years ) with a mean age of 33 ± 14·1 years . The clinical presentation and laboratory characteristics of the 357 eligible patients are summarized in Tables 3 and 4 . The mean interval from the onset of illness to the hospital presentation was 4 ± 1·8 days ( range 1–7 , median 4 days ) . One of the 357 eligible patients met the inclusion criteria despite having no fever . The inclusion criteria were any patient with clinically suspected dengue that was laboratory-confirmed with positive IgM , PCR , or NS1 and presenting within 7 days after the onset of symptoms . This patient was a 25 y old Somalian woman with one day history of generalized body ache , vomiting , menorrhagia , and dyspnea Examination revealed jaundice and splenomegaly . Platelet count was 9 . 0 x103/mm3 , WBC , 1 . 0 x103/mm3 , and hemoglobin , 4 . 6 g/dl . Dengue was clinically suspected because of her thrombocytopenia and leukopenia and the fact that she presented during an outbreak of dengue . Dengue IgM was positive . She was managed with blood transfusion and supportive care and the patient fully recovered . Since the old and the new classifications developed by the WHO were applicable to both adults and children , we decided to analyze and present the data for the whole group . Despite the small number of children ( 53 child or 15% of our study population ) , subgroup analysis was performed to see if there were any major differences between adults and children and there was none . Of 357 eligible patients , 190 ( 53% ) patients were hospitalized for a mean of 7·6 ± 9·8 days ( range 1–95 , median 5 days ) . Nineteen ( 5% ) patients required admission to the intensive care unit ( ICU ) and 8 ( 2% ) patients died . Of 357 eligible patients , 167 ( 47% ) were assessed and managed in the emergency/ambulatory departments . Cross-tabulations of the level of care ( level I/II versus level III ) with the old and the new classifications showed 86% and 93% proportional agreement ( PA ) of the old and the new classifications with the level of care , respectively ( Tables 5 and 6 ) . There was a minimal agreement ( Kappa standard error [SE] = 0·179 [0·075]; p = 0·01 ) between the two classifications , although proportional agreement ( PA ) was 85% ( p = 0·003 ) ( Table 6 ) . Diagnostic values including sensitivity , specificity , PPV and NPV of the old and the new classifications in identifying patients who required level III of care is presented in Table 7 . Of the 40 patients who needed level III of care , 14 ( 35% ) patients were not classifiable as SDF using the WHO 2009 classification . Of these 14 cases , 8 had co-existing hematological conditions ( sickle cell anemia and hereditary spherocytosis ) that caused severe drop in hemoglobin level necessitating urgent blood transfusion , 5 patients had severe thrombocytopenia ( <20 , 000 platelets/mm3 ) with evidence of bleeding , and one patient had a neurological deficit that was exacerbated by dengue . Thus , the 14 patients with SDF as per the gold standard ( ie requiring level III of care ) who were not identified by the new WHO classification were either patients with decompensated chronic medical diseases , hematological diseases presenting with aplastic crisis , or thrombocytopenia <20 , 000 platelets/mm3 and minor or major bleeding . To improve the sensitivity of the clinical assessment in identifying patients who would require level III of care , we propose two revised versions of the new classification by integrating new criteria to the definition of SDF . The first proposed revision integrates patients with acutely decompensated chronic disease; e . g . patients with known cardiomyopathy who present with acute heart failure or patients with known hematological disease , such as sickle cell anemia , presenting with aplastic crisis . This first revision improves the sensitivity from 65% to 85% , the PPV from 74% to 79% , and the NPV from 96% to 98% , with no change in specificity ( 97% ) . The second proposed revision also integrates ( besides decompensated chronic disease ) patients with thrombocytopenia of <20 , 000 platelets/mm3 with any bleeding even if minor . This second revision results in further improvement of the sensitivity from 65% to 98% , the PPV from 74% to 80% , and the NPV from 96% to 99·7% , with without affecting specificity ( 97% ) . The diagnostic value of the first and second proposed revisions of SDF in the new classification is presented in Table 8 and the revised criteria are presented in Table 9 . Severe outcomes were defined as patients who recovered after having complications directly or indirectly related to dengue or those who died . The sensitivity of the new classification , the 1st proposed revision , and the 2nd proposed revision , for identifying patients with severe outcomes was 72% , 84% , and 88% , respectively , and the NPV was 97·7% , 99% , and 99% , respectively ( Table 10 ) . Evaluation of the warning signs of the new classification showed that patients with hemoconcentration associated with concurrent drop in platelets count had approximately 5-fold increased risk of requiring level III healthcare ( OR [95% CI] = 4·97 [2·35–10·50] , p<0·001 ) and approximately 7-fold increased risk of severe outcome ( OR [95% CI] = 6·89 [2·90–16·37] , p<0·001 ) . Other warning signs were not significant predictors of level III healthcare or severe outcome ( Table 11 ) . Correlation between the number of warning signs ( <2 versus ≥2 ) , the old and the new classifications , and the level of care is presented in Table 12 .
This study compared the old and the new WHO dengue classifications as predictors of both levels of healthcare and patients' outcomes among 357 patients with confirmed dengue . It demonstrated that both classifications were inadequate in identifying patients who required advanced level of healthcare . Both classifications had a low-to-moderate sensitivity ( 30% to 65% ) , although the new classification had an acceptable sensitivity in predicting severe outcomes ( 72% ) . Consequently , we propose to revise the new classification by integrating 2 new criteria in the definition of SDF , namely , acute decompensation of chronic diseases and evidence of minor bleeding in association with thrombocytopenia <20 , 000 platelets/mm3 . The proposed revision improved the sensitivity to 98% in detecting patients who need level III healthcare without significantly altering specificity , which remained high ( 97% ) . It also improved sensitivity of predicting severe outcomes to 88% . Furthermore , the proposed revisions had 99·7% and 99% NPVs for the level of care and outcome severity , respectively . This means that a patient who would not be classified as SDF according to our proposed criteria would have 0·3% probability to require level III of healthcare and 1% probability of having severe outcome including complications and mortality; versus 5% each for the original classification , respectively . Pre-existing comorbidities have been considered as risk factors for progressing to severe dengue in the literature . In this study , patients with pre-existing morbidities that were clinically stable at presentation and throughout the duration of dengue illness did not need to be managed as severe cases ( level III of care ) . On the other hand , patients known to have chronic diseases who presented with acute decompensation ( unstable disease ) at presentation or any time during their dengue illness needed to be managed as severe dengue and received level III of care; e . g . patients with known cardiomyopathy who presented with acute heart failure or patients with known hematological disease , such as sickle cell anemia , who presented with aplastic crisis . Data from literature generally suggest that the new classification had a higher sensitivity as well as a higher or comparable specificity in identifying severe cases requiring higher level of healthcare in comparison with the old classification [14–19] . Several studies that compared the WHO classifications showed that a significant proportion of patients with SDF were misclassified as classical DF using the old classification , while they were readily identified by using SDF in the new classification [14–16] . Other studies compared the accuracy of the two classifications in certain clinical contexts . For example , a Brazilian study of 267 pediatric cases , showed that the old classification had a lower sensitivity ( 62% versus 87% ) but a higher specificity ( 93% versus 73% ) in discriminating severe cases compared to the new classification [20] . Another Brazilian retrospective study evaluating 121 autopsied individuals who died during 2011–2012 dengue epidemics , showed that the new classification had a higher sensitivity to discriminate dengue deaths and that the absence of plasma leakage and thrombocytopenia were the main reasons for failure of the old classification to discriminate DHF cases [21] . A review by Bandyopadhyay et al . , analyzed 37 studies using the old WHO dengue classification , and demonstrated that this classification had a low sensitivity with frequent overlap between the different classes , especially in endemic areas [22] . Additionally , the new classification has been shown to be a better measure for case reporting and surveillance [21–23] . Conversely , other authors found no difference between the two classifications in term of sensitivity [24 , 25] . The new classification intended to develop a system that helps in directing patients' management and improving clinical outcome by reducing morbidity and mortality . Beyond the controversy over the usefulness of the old and the new WHO definition of SDF in identifying severe cases , the practicability of the new definition in epidemic contexts was contested , as it was perceived to be entailing heavier workload to healthcare personnel [24] . According to the new classification , all patients presenting with dengue warning signs should be admitted for observation . This would raise the hospitalization rate from 10% ( DHF/DSS ) to 75% ( D+W/SDF ) according to our data , resulting in unnecessary observation/admission , which might overwhelm healthcare personnel and resources particularly during outbreaks . As shown in our study , almost half of D+W patients were managed as outpatients , which makes the relevance of warning signs in terms of disease severity questionable [14–17] . Only hemoconcentration with concurrent drop in the platelet count was a significant predictor of severe outcome and the need for advanced healthcare . Most of the previous studies found that none of the suggested warning signs was a significant predictor of dengue severity [24 , 26 , 27] , although one of them showed that the presence of five or more warning signs was a significant predictor of SDF [27] . A multi-center study across 18 countries , assessing user-friendliness and acceptance of the new classification from health professionals’ viewpoint , showed that 24% of health professionals had concerns with the new classification including: a possible increase of hospitalization rates , non-specificity of warning signs , a possible increase of cost if more patients were admitted , and the need for more training and dissemination of more concise clinical protocols [17] . These disadvantages called for a revision to improve its practicability and specificity in identifying severe cases . In the present study , the new definition of SDF missed 14 ( 35% ) patients who needed advanced ( level III ) medical care . Of these 14 cases , 8 had co-existing hematological conditions ( sickle cell anemia and hereditary spherocytosis ) that caused severe drop in hemoglobin level necessitating urgent blood transfusion , 5 patients had severe thrombocytopenia ( <20 , 000 platelets/mm3 ) with evidence of bleeding , and one patient had a neurological deficit that was exacerbated by dengue . Adding two new criteria , namely , thrombocytopenia <20 , 0000 platelets/mm3 with evidence of bleeding , even though minor , and decompensated chronic illness , to the definition of SDF improved its sensitivity to identify patients who needed advanced level of care from 65% to 98% and its sensitivity to predict cases likely to have morbidity or mortality from 72% to 88% . The first criterion suggested to be added to our proposed revision of the new classification as an indication of SDF is “acute decompensation of a preexisting comorbidity” . Hematological conditions constituted majority of the cases where 8/10 patients who had a co-existing sickle cell anemia or hereditary spherocytosis developed aplastic crisis . Dengue is a hemorrhagic virus and it’s known to cause thrombocytopenia and leukopenia but not anemia . However , a pre-existing defect along the line of red blood cells potentiated acute anemia . The second criterion suggested to be included in our proposed revision of the new SDF definition is “thrombocytopenia<20 , 000 platelets/mm3 with any bleeding , even though minor” . Thrombocytopenia ≤100 , 000 platelets/mm3 constitutes only one of the warning signs in the original version of the new WHO classification ( 2009 ) but it is not included in the definition of SDF . In a French-Polynesian study , thrombocytopenia <20 , 000 platelets/mm3 was associated with a longer hospital stay , more frequent admission to ICUs , and higher mortality [28] . However , only two thirds of cases were classified as DHF/DSS and the remaining one third was classified as DF resulting in underestimation of the severity of illness [28] . This is consistent with our results demonstrating that severe thrombocytopenia ( <20 , 000 platelets/mm3 ) is a strong indicator for severe dengue and the need for advanced level of care . Therefore , the new classification should be revised to include severe thrombocytopenia in the definition of SDF to improve identification of severe cases . In our study , 9/35 patients classified as SDF received level II of healthcare and the remaining 26 patients were hospitalized for close observation and monitoring . Of these 26 patients , 3 patients had suspicion of liver damage ( aminotransferases >1000 U/L ) , which proved to be a transient elevation of liver enzymes resolving spontaneously without complications . Among the patients who had neurological manifestations , impaired consciousness was the most frequent sign ( 5% ) , followed by meningeal signs ( 5% ) , and convulsion ( 3% ) with one patient presenting with encephalitis . All cases with neurological manifestations benefited from conservative management . Similar to our findings , a Vietnamese study reported that impaired consciousness was the most frequent neurological manifestation [29] . Impaired consciousness is part of SDF criteria according to the new classification . A study by Gupta et al , demonstrated a greater risk for neurological complications among patients classified as classical DF in the old classification which did not include neurological symptoms or low level of consciousness at presentation as signs of SDF [30] . Recent reviews also highlighted a rising trend of neurological complications of dengue and that it is associated with more frequent hemorrhagic manifestations , higher prevalence of DSS , and increased hospital stay and mortality [31] . A study from Brazil reported 21% of neurological manifestations with confusion being the most frequent sign [32] . In Europe , an even higher prevalence ( 24% ) of neurological manifestations was reported among imported cases of dengue infection , mainly including cases acquired in Asia and the Americas [33] . This variability in the prevalence of neurological manifestations may be explained by varying time of clinical assessment with respect to patient’s first presentation and symptoms onset . Furthermore , lack of clear definitions of organ damage that reflects the actual disease severity may lead to misplacement of patients under SDF category . In our cohort , only 61% of patients having organ damage needed level III care . Thus , standardized definition of organ damage is needed to improve the classification specificity . Having a more sensitive classification will help capture severe cases and place patients in the right interventional care level they need to improve their quality of care and subsequently decrease morbidity and mortality associated with dengue . Having a firm classification will direct patient management as per guidelines instead of relying on individualized decisions and experiences in directing patient care . Additionally , having a more sensitive classification without altering specificity will help direct healthcare resources and avoid unnecessary hospitalization/observation which usually overwhelms health care systems especially during outbreaks . Limitations of our study include the retrospective design yielding incomplete observations , especially for cases treated on outpatient basis , which may have compromised the robustness of the analysis . Moreover , most of the severe cases presented at a late stage , which limited the accuracy of warning signs assessment and analysis as predictors for severe dengue . Furthermore , this is a limited study in a specific area and transferability to other countries is questionable . In conclusion , the new WHO dengue classification had low sensitivity for identifying patients in need of advanced level of care and for predicting morbidity and mortality . This classification needs to be revised to improve its sensitivity in predicting the required level of care . It is proposed to include 2 additional criteria to the definition of SDF , namely , decompensation of chronic disease and thrombocytopenia <20 , 000 platelets/mm3 with evidence of any bleeding , even though minor . Adding these two criteria substantially improves the sensitivity to predict cases requiring advanced level of care and to predict severe outcomes . Further prospective controlled two-arm study designs are needed from more than one region and different countries to confirm our results . Further , criteria such as organ damage and impaired consciousness should be distinctively defined to avoid overlap of the definitions and misclassification . | Dengue fever , the most prevalent arthropod-borne viral disease in humans , has been conventionally classified into four main categories: non-classical , classical , dengue hemorrhagic fever , and dengue shock syndrome . Several studies reported lack of correlation between the categories of the conventional classification and the disease severity . As a consequence , the World Health Organization proposed in 2008 a new classification that divides dengue into two categories: non-severe and severe dengue; the non-severe dengue is further divided into two categories: dengue with warning signs and dengue without warning signs . In this retrospective study we reviewed 357 cases of dengue diagnosed in our institution over a 4-year period to assess the validity of the new dengue classification in order to develop pragmatic guidelines for case triage and management in the Emergency Departments . We found that the sensitivity of the new classification for identifying severe cases was limited even though it had a higher sensitivity for identifying patients who needed advanced healthcare compared to the old one . We propose adding decompensation of chronic diseases and low platelets-related bleeding to the category of severe dengue in the new classification . This modification dramatically improves the sensitivity for identifying patients who need advanced healthcare and the sensitivity to predict severe outcomes . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"cognitive",
"science",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"vomiting",
"tropical",
"diseases",
"neuroscience",
"physiological",
"processes",
"cognitive",
"neuroscience",
"signs",
"an... | 2019 | Assessment of the new World Health Organization's dengue classification for predicting severity of illness and level of healthcare required |
Among the most common human congenital anomalies , cleft lip and palate ( CL/P ) affects up to 1 in 700 live births . MicroRNA ( miR ) s are small , non-coding RNAs that repress gene expression post-transcriptionally . The miR-17-92 cluster encodes six miRs that have been implicated in human cancers and heart development . We discovered that miR-17-92 mutant embryos had severe craniofacial phenotypes , including incompletely penetrant CL/P and mandibular hypoplasia . Embryos that were compound mutant for miR-17-92 and the related miR-106b-25 cluster had completely penetrant CL/P . Expression of Tbx1 and Tbx3 , the DiGeorge/velo-cardio-facial ( DGS ) and Ulnar-mammary syndrome ( UMS ) disease genes , was expanded in miR-17-92 mutant craniofacial structures . Both Tbx1 and Tbx3 had functional miR seed sequences that mediated gene repression . Analysis of miR-17-92 regulatory regions uncovered conserved and functional AP-2α recognition elements that directed miR-17-92 expression . Together , our data indicate that miR-17-92 modulates expression of critical T-box transcriptional regulators during midface development and is itself a target of Bmp-signaling and the craniofacial pioneer factor AP-2α . Our data are the first genetic evidence that an individual miR or miR cluster is functionally important in mammalian CL/P .
The evidence that there is a genetic component underlying CL/P is compelling . Analysis of a Danish cohort of CL/P cases revealed that relatives of patients with CL/P have a higher relative risk for CL/P compared to background risk levels . This notion of CL/P heritability is also supported by twin studies [1] , [2] . Genome wide association studies ( GWAS ) and mouse genetics studies have also pointed to genes and genomic regions that are associated with CL/P [3] , [4] . MiRs repress gene expression post-transcriptionally by Watson-Crick base pairing to the seed sites in the 3′UTR of target genes . The miR-17-92 cluster , encoding miR-17 , miR-18a , miR-19a , miR-20a , miR-19b-1 , and miR-92a-1 , is within a region on chromosome 13q that when deletion is associated with CL/P , lung hypoplasia , microphthalmia , microcephaly , and small stature in human patients and has phenotypic similarities to Feingold syndrome [5] , [6] . Moreover , miR-17-92 is found in an amplified region associated with small cell lung cancer , as well as in B-cell lymphomas , and is over-expressed in several solid tumor types , including breast , colon , lung , pancreas , and prostate cancers [7] . The mouse embryos with miR-17-92 loss-of-function have smaller body size , microphthalmia , heart defects , and lung hypoplasia [6] , [8] , [9] . Moreover , the miR-17-92 gain-of-function mutants develop lymphoma , indicating that the mouse is an accurate model for the human syndrome [10] . Unlike the miR-17-92 loss-of-function mice , its two homologous clusters , miR-106a-363 and miR-106b-25 loss-of-function embryos do not exhibit any gross abnormalities . Our previous findings indicated that miR-17-92 is directly regulated by Bmp-signaling in heart development [9] . Bmp-signaling deficiency in mice and humans has been shown to cause CL/P and other craniofacial anomalies [11] , [12] . Interestingly , miR-17-92 has also been shown to be directly regulated by Myc family transcription factors [7] . Here , we show that miR-17-92 deficiency results in orofacial clefting and that the human disease genes Tbx1 and Tbx3 are direct targets for miR-17-92 . Our findings also reveal that miR-17-92 is a direct target for the master regulator of cranial neural crest development AP-2α .
We found that miR-17-92 ( miR17-92null/null ) mutant embryos had severe craniofacial defects including CL/P and mandibular hypoplasia with notching , revealing that miR-17-92 is a critical regulator of craniofacial development ( Figure 1A–H , Figure S1 ) . Moreover , by genetically reducing miR-106b-25 dose on the miR-17-92null background , the clefting phenotype was both more severe and completely penetrant , indicating that there is genetic redundancy between these two miR complexes ( Figure 1C–F , Figure S1E–H and Table S1 ) . In addition to cleft lip and mandible defects , both miR-17-92 mutants and miR-17-92null; miR-106b-25null compound mutants had cleft secondary palate ( Figure 1 B , D , H ) . Expression of mitotic cell marker phospho-Histone H3 ( pHH3 ) was greatly reduced in miR-17-92 mutants , indicating that miR-17-92 is required for normal progenitor cell proliferation during orofacial development ( Figure 1I–L , Figure S2 ) . Taken together , these data provide the first genetic evidence that miRs are important regulators of mammalian orofacial development and are involved in CL/P . We generated a miR-17-92 bacterial artificial chromosome ( BAC ) transgenic LacZ reporter line to follow the expression of primary ( pri ) -miR-17-92 ( Figure S3A ) . Three individual transgenic lines showed similar LacZ expression pattern , revealing that pri-miR-17-92 was expressed in branchial arches and frontonasal process ( Figure 2A ) . LacZ was also detected in the nasal structures , calvarial bones , auricle , periocular mesenchyme , and limb mesenchyme ( Figure 2K ) . Sagittal sections on E11 . 5 embryos revealed LacZ activity in epithelium and mesenchyme of first branchial arch and frontonasal process ( Figure 2I ) . Coronal sections through E12 . 5 and E13 . 5 embryos demonstrated LacZ staining in distal tips of the palatal shelves , the mandibular mesenchyme and mesenchyme of forming frontal bones ( Figure 2J , L ) . In situ with a pri-miR-17-92 probe revealed similar expression pattern as the transgenic LacZ data ( Figure 2E–F ) . Furthermore , in situ analysis with locked nuclei acid ( LNA ) probes to detect mature miR-17 and miR-92a showed that miR-17 and miR-92a were highly expressed in branchial arch and frontonasal process ( Figure 2 B–D , G–H ) . Unlike miR-17-92 , expression of miR-106b-25 was relatively low ( Figure S3B–E ) . Tbx1 gain-of-function causes cleft lip and is a miR-17-92 target in the heart [9] , [13] . We evaluated the expression of candidate craniofacial miR-17-92 target genes based on bioinformatics analysis , including Tbx3 , Fgf10 , Pax9 , Shox2 and Osr1 , in both miR-17-92 null and conditional knock out mutants using AP-2α cre driver [14] . In situ hybridization in miR-17-92 mutants demonstrated up-regulated Tbx3 expression in mandible , frontonasal-derived structures , tongue , and secondary palate at E13 . 5 ( Figure S4A–F , Figure S5E–F ) and in paired maxillary processes and nasal process at E10 . 0 and E10 . 5 ( Figure 3A–B , Figure S4 G–N ) . Changes in Tbx1 and Fgf10 expression were not detected at E10 . 0 likely because the expression changes were not dramatic enough to be detected by in situ hybridization ( data not shown ) . Tbx1 was expanded primarily in the secondary palate , tongue , and oral ectoderm at E13 . 5 ( Figure 3C–D , Figure S5A–D ) . In addition , Fgf10 was expanded in distal mandible and tongue ( Figure S5I–L ) , while ectopic Shox2 expression was observed in distal mandible at E13 . 5 ( Figure S5G–H ) . Expression of Osr1 was upregulated in the distal mandible and frontonasal structures at E13 . 5 ( Figure S5M–P ) . In contrast , the expression pattern of Pax9 in miR-17-92 null mutant embryos was unchanged compared to control embryos ( data not shown ) . qRT-PCR experiments also showed up-regulation of Tbx1 , Tbx3 , Fgf10 , Shox2 and Osr1 in miR-17-92; miR106b-25 compound mutants at E13 . 5 ( Figure 3E ) . To evaluate the expression changes of the above genes , we used a miR-17-92 conditional , cre-activated gain-of-function line ( miR-17-92OE ) and the Wnt1cre driver to activate miR-17-92 in cranial neural crest ( CNC ) [10] , [15] . qRT-PCR analysis from Wnt1cre; miR-17-92OE orofacial tissue revealed that Fgf10 , Tbx1 , Tbx3 , Osr1 and Shox2 were significantly repressed ( Figure 3F ) , while there was no obvious morphological defect detected in miR-17-92 overexpression mutants potentially due to moderaterepression of the miR-17-92 target genes . Target genes that are repressed by miR-17-92 have a mixture of miR-17/20a/106b and miR-92a/25 family seed sites in their 3′UTRs ( Figure S6 ) . Bioinformatics analysis revealed conserved miR-17/20a/106b family seed sequence in the 3′ UTR of Fgf10 , Shox2 and Osr1 ( Figure S6A–C , F ) . The Tbx3 3′ UTR contained both a miR-17/20a/106b family seed site and a miR-92a/25 family seed site ( Figure S6D–E ) . We cloned the 3′ UTRs of Fgf10 , Shox2 , Tbx3 and Osr1 into luc reporter plasmids to test miR seed sequence function in vitro . Transfections with miR mimics of miR-17-92 resulted in drastic reduction in luciferase activity for all of the reporter plasmids ( Figure 3G , Figure S7 ) . Mutation of the respective miR seed sequences within 3′ UTRs of those genes ablated the inhibition by the corresponding miR ( Figure 3G , Figure S7 ) . These data suggest that miR-17-92 directly inhibits Fgf10 , Shox2 , Tbx3 and Osr1 . Previous work showed that conditional inactivation of Bmpr1a , Bmp4 , and Bmp2;Bmp4 in developing facial processes using the Nestincre transgenic driver result in orofacial clefting ( [11] and Figure S8 ) . This cre driver directs cre activity in facial prominences [11] . Moreover , miR-17-92 is a direct target for Bmp-signaling in cardiac progenitors [9] . We crossed the miR-17-92OE line into Nestincre , Bmp4 , Bmp7 conditional mutant background to test whether miR-17-92 gain-of-function could genetically rescue the defects in Bmp mutants . All NestinCre , Bmp4 flox/+ , Bmp7 flox/+ embryos ( 23 out of 23 ) and embryos without NestinCre ( 29 out of 29 ) had normal morphology ( Figure 4A , Figure S9A , D , table S2 ) , while all NestinCre , Bmp4 flox/flox , Bmp7 flox/+ mutant embryos ( 6 out of 6 ) had bi-lateral cleft lip and heart defects with incompletely penetrant embryonic lethality at E12 . 0 likely due to heart defects ( Figure 4B , Figure S9B , E , table S2 ) . Most ( 5 out of 6 ) NestinCre , Bmp4 flox/flox , Bmp7 flox/+ , miR-17-92OE embryos were rescued by miR-17-92 overexpression ( significant different compared to NestinCre , Bmp4 flox/flox , Bmp7 flox/+ mutants , CHI-TEST , p<0 . 01 ) , with full suppression of cleft lip and heart defect caused by Bmp loss-of-function , but not eye defect ( Figure 4C , Figure S9C , F , table S2 ) . Consistently , qRT-PCR data indicated that pri-miR-17-92 , miR-17 , and miR-20a were reduced in Bmp2/4 mutant midface ( Figure 4D ) . In situ analysis using miR-17 LNA probe also indicated that miR-17 was dramatically reduced in Bmp2/4 mutants ( Figure S10C , D ) compared to controls ( Figure S10A , B ) . In addition , qRT-PCR indicated that Fgf10 , Tbx1 , Tbx3 , Osr1 and Shox2 were up-regulated in the midface of Bmp2; Bmp4 conditional mutants ( Figure S10E ) , further suggesting that these genes are regulated by a BMP-miR-17-92 genetic pathway in craniofacial structures . Moreover , in vivo chromatin immunoprecipitation ( ChIP ) data using embryonic midface extracts showed enrichment in the anti-Smad1/5/8 immunoprecipitated chromatin , indicating that Smad1/5/8 directly binds miR-17-92 chromatin ( Figure 4E ) . Co-transfection of a constitutively active Bmpr1a construct with miR-17-92 luc reporter resulted in approximately 3-fold induction supporting the hypothesis that Bmp signaling directly regulates miR-17-92 in developing craniofacial structures ( Figure 4F ) . Together , a conserved Bmp-miR-17-92 genetic pathway plays a critical role in the orofacial development . Mouse mutants for AP-2α have CL/P and human patients have branchio-oculo-facial syndrome that has CL/P as a cardinal feature ( BOFS MIM 113620 ) . ChIP-sequencing ( ChIP-seq ) indicated that AP-2α bound to miR-17-92 chromatin in cultured human neural crest [16] ( Figure 4H , S11A ) . To determine if AP-2α directly regulates miR-17-92 , we evaluated pri-miR-17-92 , mature miR-17 , and mature miR-20a levels in the AP-2α mutant midface . qRT-PCR experiments indicated that pri-miR-17-92 and mature miRs were down-regulated in AP-2α mutants ( Figure 4G ) . We used ChIP-PCR to determine whether AP-2α binds to miR-17-92 chromatin in developing midface tissue . Because there are multiple predicted AP-2α binding sites in miR-17-92 , we subdivided miR-17-92 into four regions based on ChIP-seq ( Figure 4 H–J and S11 ) . ChIP-PCR experiments using midface extracts indicated that AP-2α bound to miR-17-92 regions 1 , region 2 , and region 4 ( Figure 4 K ) . Transfection experiments with a miR-17-92 reporter containing AP-2α binding sites revealed that AP-2α transcriptionally activated miR-17-92 although synergism with Smad1 was not detected using this miR-17-92 reporter ( Figure 4 L ) . Moreover , AP-2α may also directly regulate miR-106b-25 as suggested by the analysis of ChIP-seq data [16] ( Figure S12 ) .
Tbx1 loss- and gain-of-function result in cleft palate in human DGS patients and mouse models [13] , [17]–[19] . Consistent with our findings , Tbx1 gain-of-function results in cell cycle arrest [17] . DGS is characterized by highly variable phenotypes indicating that there are strong modifiers in the human genome [18] , [20] . Our findings suggest miR-17-92 as a candidate genetic modifier for Tbx1 since it fine-tunes Tbx1 expression levels . Mouse mutants for Tbx3 and the related Tbx2 have cleft palate [21] . Furthermore , human patients with UMS have abnormal and distinct facial appearance indicating a requirement for Tbx3 in human craniofacial development [22] . While our findings suggest that elevated Tbx3 inhibits proliferation , there is other evidence suggesting that Tbx3 promotes proliferation [23] . However , an in vivo study reveals that Tbx3 overexpression results in reduced cardiomyocyte proliferation in the zebrafish heart [24] . More work will be required to evaluate Tbx3 function and target genes in vivo in the context of the miR-17-92 mutant midface to better understand contextual Tbx3 function . Both Fgf10 and Fgfr null mice have cleft secondary palate [25] , [26] . Mutations in Fgf10 and Fgf receptors cause lacrimo-auricular-dento-digital ( LADD ) syndrome in human patients indicating a requirement for Fgf-signaling in human craniofacial development [27] . Fgf10 mRNA is enriched in anterior and middle regions of the secondary palate . Moreover , Fgf10 deficiency results in abnormal fusion of the palatal to oral cavity epithelium , suggesting that Fgf10 is required for maturation of palate epithelium . Importantly , elevated Fgf signaling is pathologic in human patients as shown by the extensive investigations into Fgf receptor mediated craniosynostosis [28] . Homozygosity for the Fgfr2 gain-of-function Crouzon mutation in mice results in cleft palate , as well as , craniosynostosis [29] indicating that elevated Fgf signaling also causes cleft palate . Our data demonstrate that miR-17-92 directly represses Fgf10 as a mechanism to maintain correct levels of Fgf10 during palate closure . Currently , there are no other genetic loss-of-function data indicating that single miRs or miR clusters are important in mammalian orofacial clefting . Data from zebrafish indicate that miR-140 targets pdgfra to regulate primary palate development [30] . GWAS in human patients reveal important genome regions that are associated with CL/P , including 8q24 [3] , [31] . Within the 8q24 region is the c-myc gene , a known miR-17-92 regulator [32] , [33] . Chromosomal deletions that include miR-17-92 cause a variant of Feingold syndrome in human patients with small stature and skeletal abnormalities [6] . Human patients with hemizygous miR-17-92 deletion do not have CL/P likely reflecting phenotypic heterogeneity in miR-17-92 loss of function families . These data are consistent with our findings indicating that there is incomplete penetrance of the CL/P phenotype in miR-17-92 mutant mouse embryos ( table S1 ) . Consistent with previously finding that Bmp-deficiency results in CL/P in mice and humans [11] , [12] , our data indicate that Bmp signaling activates miR-17-92 in craniofacial development . Moreover , we show that AP-2α also regulates miR-17-92 expression although our transfection assays failed to uncover synergistic miR-17-92 activation by AP-2α and Bmp-signaling ( not shown ) . One possibility is that Bmp-signaling and AP-2α activate miR-17-92 sequentially during craniofacial progenitor cell development . The assays we employed here cannot easily distinguish molecular events that occur in neighboring or closely apposed cells rather than in the same cell . We also failed to detect up-regulated miR-17-92 target genes in AP-2α mutants perhaps due to functional redundancy with other AP-2 family members [34]–[36] . Nonetheless our findings have important implications since AP-2α has been shown to regulate Irf6 , a common genetic defect in syndromic and non-syndromic CL/P in human patients [37] . AP-2α regulation potentially connects miR-17-92 to a gene regulatory network that may be involved in a large portion of human CL/P . In summary , we identified a miR-mediated genetic pathway that plays critical roles during orofacial development ( Figure S13 ) .
All animal experiments detailed within the manuscript were approved by the Baylor College of Medicine review board . The miR-17-92 and miR-106b-25 alleles , Bmp2 , Bmp4 and Bmp7 conditional null , AP-2α cre , Nestin cre and Wnt1cre alleles were previously described [8] , [10] , [11] , [14] , [15] , [38] . To generate miR-17-92-lacZ reporter transgenic lines , we obtained the BAC from BACPAC Resources Center , Children's hospital Oakland Research Institute ( BAC number: RP23-89P9 ) and replaced miR-17-92 sequence with lacZ coding sequence by recombineering , followed by pro-nuclear injection . Constructs were generated using PCR , cloning and recombineering . A LoxP site flanked neo cassette was isolated from PL452 plasmid using BamHI and EcoRI . lacZ coding sequence was isolated from hsp68-lacZ plasmid using BamHI and NcoI . All fragments were cloned into pBluescript SK+ to generate miR-17-92-lacZ construct , followed by recombineering and subsequently Cre mediated recombination for removal of the neo cassette ( Figure S3A ) . Mouse embryos were harvested in ice-cold Phosphate Buffered Saline ( PBS ) , then fixed overnight ( O/N ) in 4% paraformaldehyde ( PFA ) and 2% glutaraldehyde in PBS at 4°C . The samples were then dehydrated in ethanol series to a final 100% ethanol , followed by transferring to graded series of increasing concentrations of hexamethyldisilazane ( HMDS ) for 5 min each and air dried O/N . Samples were mounted on to double-stick carbon tabs ( Ted Pella . Inc . ) , which have been pre-mounted on to aluminum specimen mounts ( Electron Microscopy Sciences ) . The samples were then coated with a thickness of 25 nm platinum alloy under vacuum using a Balzer MED 010 evaporator ( Technotrade International ) , then immediately flash carbon coated under the same vacuum . The samples were transferred to a desiccator for later examination . JSM-5910 scanning electron microscope ( JEOL , USA , Inc . ) was used at an accelerating voltage of 5 kV . Embryos were fixed in 4% PFA , embedded in paraffin and cut to 5 µm sections mounted on Superfrost/Plus slides ( Fisher Scientific ) . The antigens were retrieved by incubating in the citrate buffer ( 10 mM ) for 2 minutes in microwave oven . The primary antibody was anti-Phospho-Histone H3 with 1∶200 dilution ( Cell Signaling ) . Broad HRP conjugated secondary antibody ( Invitrogen ) was used and visualized using TSA Plus Fluorescence Systems from PerkinElmer on a Zeiss LSM 510 Confocal Microscope . Nuclei were stained with 4 , 6-diamidino-2-phenylindole ( DAPI ) . Tissue preparation and in situ hybridization were as previously described [39] , [40] . The gene probes were synthesized using DIG RNA Labeling Kit ( Roche ) following manufacturer's guidelines . The enzymes used for digestion and transcription of in situ constructs are SacII and T7 for Fgf10 , XhoI and T7 for Shox2 ( gift from Dr . Yiping Chen's lab ) , EcoRI and T7 for Osr1 ( gift from Dr . Rulang Jiang's lab ) , EcoRI and T3 for Tbx1 ( gift from Dr . Antonio Baldini's lab ) , PstI and T3 for Tbx3 ( gift from Dr . Robert Kelly's lab ) . miRCURY LNA probes were purchased from Exiqon and used per manufacturer's guidelines . To generate 3′ UTR luciferase reporter plasmids , 3′ UTR genomic sequence of genes including Fgf10 , Osr1 , Shox2 and Tbx3 were amplified using a high-fidelity PCR system ( Roche ) with designed oligonucleotides and subcloned into the pMIR-REPORT Luciferase miRNA Expression Reporter Vector ( Ambion ) . Oligonucleotides used to amplify 3′ UTR genomic sequence of Fgf10 are sense , 5′-CGACTAGTAAGAAAACACTGTTGGTGGATGCAG -3′ , and antisense , 5′-GCACGCGTTTTTATTCTCTTTTCCCAGC-3′ . Oligonucleotides used to amplify 3′ UTR genomic sequence of Osr1 are sense , 5′- GACTAGTATAAACAGAGCCTGCGGG -3′ , and antisense , 5′- CGACGCGTGCCTGTAAAATAACCGTTTATTT -3′ . Oligonucleotides used to amplify 3′ UTR genomic sequence of Shox2 are sense , 5′-ACTAGTCGCCGGCGCCAGCGCCACGGT-3′ , and antisense , 5′-AAGCTTCTTTTTTGTATGAACGTCC-3′ . Oligonucleotides used to amplify 3′ UTR genomic sequence of Tbx3 are sense , 5′- GACTAGTAAACAAGAAAAACAAAATCGCC -3′ , and antisense , 5′- CCCAAGCTTTCATTTCAATAAAAATTTATTG -3′ . Oligonucleotides used to amplify 3′ UTR genomic sequence of Tbx3 without mir17/mir-20a seed site are sense , 5′- GACTAGTGTGTAACCAGGCTGCTGTTGCTTT -3′ , and antisense , 5′- CCCAAGCTTTGGTCGTTTGAACCAAGTCCCTCT -3′ . Underlined letters represent enzyme restriction sites for subcloning . All PCR products were sequenced to make sure no mutations were introduced . All site-directed mutagenesis of the miR seed sites in the 3′ UTR reporter constructs were achieved by using the QuikChange II site-directed mutagenesis kit ( Stratagene ) . The sense-strand sequences of the oligonucleotides used for mutagenesis ( underlined letters indicate the mutation of miR seed sites ) were: 5′-TAAGACACGCAAGCATTTACTGGAAAGACACTGGGTCATATCATATGCACAACCAAAG- 3′ ( Fgf10-mut1 , ) , 5′-CCCCATGCGCTCTCAGTTGACTTAATTTGACACTCTGCAATAAAAAACACCAGCAAT- 3′ ( Fgf10-mut2 ) , 5′-ACAGCAAATAGTGCAGACGTTGGATTCTTATTTCAACCCGCCATTTAGATTACTAAAGAGA- 3′ ( Fgf10-mut3 ) ; 5′-GCTGACCTTTTTCTGCGAAGTTGAATTCAATAGGAGACATTTGATAAGAG - 3′ ( Shox2-mut ) ; 5′- GCCGGGCGTTGTATTGCGACTGGGAATTCATGCTGACCATCGGTAACGGAC - 3′ ( Osr1-mut ) ; 5′- GGACCATTAGTTCTTTTAACTGTATAGAATTCAACAAGGTTTTAAAAGATAATAATA - 3′ ( Tbx3-mut ) . All PCR products were sequenced to make sure no unexpected mutations were introduced . Wild type mouse embryonic orofaces were dissected at E12 . 5 ( for Smad1/5/8 ChIP ) or E10 . 5 ( for AP-2α ChIP ) and followed by ChIP analysis as previously described [9] . As control , normal rabbit immunoglobulin G was used as a replacement for the anti-Smad1/5/8 ( sc-6031 X , Santa Cruz ) and 3B5 mouse monoclonal AP-2α antibody [41] to reveal nonspecific immunoprecipitation of the chromatin . The PCR products were evaluated for appropriate size on a 2% agarose gel and were confirmed by sequencing . The primers for amplifying the regulatory element in the 5′ upstream of mir-17-92 genomic sequence were: sense , 5′- CTGGCGGGAAGCCTGAGC -3′ , antisense , 5′-CACGGCGGCTCGTTCTTG -3′ ( for AP-2α region1 ) ; sense , 5′- CCTTCATTCACCCACATGGTCCTT -3′ , antisense , 5′- AGCAGCCGCCACCATCTT -3′ ( for AP-2α region2 ) ; sense , 5′- GCACACAATGGCCCTCGG -3′ , antisense , 5′- GCGCGCACAAAGTTTCGG -3′ ( for AP-2α region3 ) ; sense , 5′- CGCAGCCGCCCAGAAAC -3′ , and antisense , 5′-TCCGCGCCAGCTTATCAAGAGAAA -3′ ( for AP-2α region4 and Bmp/Smad regulatory element ) . Total RNA was isolated using RNeasy Micro Kit ( QIAGEN ) and real-time thermal cycling was performed using StepOne Real-Time PCR Systems ( Applied Biosystems ) . Super Script II Reverse Transcriptase ( Invitrogen ) was used for RT-PCR and SYBR Green JumpStart Taq ReadyMix ( SIGMA ) was used for real-time thermal cycling . All error bars represent SEM . Plasmids used for transfection were generated as described above or previously reported [9] . LS8 cells were transfected using Lipofectamine 2000 ( Invitrogen ) . Luciferase activity assays were measured using the luciferase Assay System ( Promega ) . hNCC AP-2α ChIP-seq and histone modification markers ChIP-seq datasets were accessed from GEO under accession number GSE28876 [16] , [42] . Raw fastq reads were mapped to hg19 genome using Bowtie2 [43] . The total number of tags of each ChIP-seq run was normalized to 10 million . ChIP-seq tracks were visualized and compared in UCSC Genome Browser . | CL/P are very common birth defects in humans . The genetic mechanism underlying CL/P pathogenesis is poorly understood . MiRs , small non-coding RNAs that function to post-transcriptionally regulate gene expression , have been identified as pivotal modulators of various developmental events and diseases . To date , there is no individual miR or miR cluster that has been identified as functionally essential in mammalian CL/P . Here , we have discovered that deletion of miR-17-92 cluster in mice results in craniofacial malformations including CL/P . Importantly , MIR-17-92 is located on a critical human chromosome region associated with 13q deletion syndrome , a chromosomal disorder that presents with defects including CL/P , suggesting the advantages of our animal model to study human disease . Moreover , our work demonstrated that miR-17-92 cluster directly repressed T-box factors , which have critical functions during craniofacial development . We further showed that miR-17-92 was directly activated by Bmp-signaling and transcription factor AP-2α . Together , our work identified a novel miR-mediated transcriptional network underlying CL/P , providing new insights into craniofacial developmental biology . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | MicroRNA-17-92, a Direct Ap-2α Transcriptional Target, Modulates T-Box Factor Activity in Orofacial Clefting |
Anthropogenic land use may influence transmission of multi-host vector-borne pathogens by changing diversity , relative abundance , and community composition of reservoir hosts . These reservoir hosts may have varying competence for vector-borne pathogens depending on species-specific characteristics , such as life history strategy . The objective of this study is to evaluate how anthropogenic land use change influences blood meal species composition and the effects of changing blood meal species composition on the parasite infection rate of the Chagas disease vector Rhodnius pallescens in Panama . R . pallescens vectors ( N = 643 ) were collected in different habitat types across a gradient of anthropogenic disturbance . Blood meal species in DNA extracted from these vectors was identified in 243 ( 40 . 3% ) vectors by amplification and sequencing of a vertebrate-specific fragment of the 12SrRNA gene , and T . cruzi vector infection was determined by pcr . Vector infection rate was significantly greater in deforested habitats as compared to contiguous forests . Forty-two different species of blood meal were identified in R . pallescens , and species composition of blood meals varied across habitat types . Mammals ( 88 . 3% ) dominated R . pallescens blood meals . Xenarthrans ( sloths and tamanduas ) were the most frequently identified species in blood meals across all habitat types . A regression tree analysis indicated that blood meal species diversity , host life history strategy ( measured as rmax , the maximum intrinsic rate of population increase ) , and habitat type ( forest fragments and peridomiciliary sites ) were important determinants of vector infection with T . cruzi . The mean intrinsic rate of increase and the skewness and variability of rmax were positively associated with higher vector infection rate at a site . In this study , anthropogenic landscape disturbance increased vector infection with T . cruzi , potentially by changing host community structure to favor hosts that are short-lived with high reproductive rates . Study results apply to potential environmental management strategies for Chagas disease .
Changes in reservoir host diversity and composition are potential drivers of vector-borne disease transmission in changing landscapes . The ‘dilution effect’ hypothesis , a widely studied biodiversity-disease related idea [1] , [2] , states that as species diversity increases , infectious disease risk decreases . The mechanism for this inverse relationship between diversity and infectious disease transmission shares the assumptions of zooprophylaxis , a principle stated earlier by public health entomologists [3] , where heterogeneities in host species competence for pathogen development and a host density-dependent , non-selective , foraging by vectors could lead to the observed patterns [4] . In theory , the ‘dilution effect’ is supported under the following conditions: variation in competence between different host species to transmit an infectious agent with greater within than between species disease transmission , dominance of the most competent reservoir in cases of species diversity , frequency-dependent transmission ( e . g . vector-borne disease ) , or density dependent transmission where an adding an additional host species decreases the relative importance of the primary reservoir host [1] , [2] , [5]–[7] . Other mechanisms by which adding species decreases disease transmission include diminishing encounters between a key host species and/or a vector , decreased efficiency of transmission with an additional host , increase in mortality of susceptible hosts , increased recovery rates of infected hosts , and decline in the number of susceptible hosts [1] , [8] . Although a growing number studies have reported inverse relationships between host diversity and infectious disease transmission or prevalence in a variety of disease systems [9]–[16] , the underlying mechanisms accounting for these observations require further study . Anthropogenic land use change may also influence vector-borne disease transmission by altering host community structure and trophic complexity . Habitat fragmentation and loss may lead to a loss of important trophic components of ecosystems , such as top-level predators , and cause relative increases in a particular species' density [17] . In disturbed landscapes , ‘mesopredator release’ [18] , [19] may occur , causing medium sized opportunistic carnivores-omnivores to increase in abundance . Frequently , these rapidly reproducing mesopredators ( e . g . opossums , raccoons ) are important reservoirs for vector-borne zoonotic diseases [20]–[26] . This is extremely interesting , because disturbance-induced biodiversity changes imply that there will also be differences in the life history of the potential hosts for a zoonotic disease . Life history changes in the predominant hosts of a zoonotic disease could drive concomitant changes in the vertebrate intra-host parasite population dynamics , and also in the transmission cycle as a whole . Based on life history theory , long-lived species with a slow reproductive rate and long lifespan should invest more resources in acquired immunity than short-lived species with high reproductive rates [27] . Additionally , those reservoir hosts that have high reproductive rates will be introducing more susceptible individuals into the host community at a higher rate than the ‘slow-living’ species . This increased number of susceptible reservoirs in a host population may allow for increased generalist vector-borne pathogen transmission . Therefore , anthropogenically disturbed habitats with a predominance of ‘fast-living’ or ‘r’ selected species compared to ecologically ‘intact’ areas [28] , [29] should be sites of increased vector-borne disease transmission within a landscape . For example , experimental studies in an amphibian-trematode systems support the hypothesis that ‘fast-living’ species are more susceptible to trematode infection and pathogenic effects [30] . An interesting system to test hypothesis regarding the role of species diversity , land use change and the heterogenity of host species life history is American trypanosomiasis , a . k . a . Chagas disease , whose etiologic agent is Trypanosoma cruzi , a vector-borne kinetoplastid protozoan parasite . Throughout Latin America , infection with T . cruzi causes significant morbidity and mortality , with estimates of 10 to over 15 million infected people [31] . T . cruzi is transmitted between a number of different mammalian reservoir hosts ( with over 100 potential host species identified ) by hematophagous vectors in the family Reduviidae [32] . In the area of the Panama Canal , T . cruzi poses a significant risk to human health and is primarily transmitted by the triatomine vector Rhodnius pallescens [33]–[36] . Data support that R . pallescens is a generalist vector in Panama [35]–[37] . R . pallescens is associated with ‘sylvatic’ habitats ( as opposed to domestic areas such as houses ) , and lives and reproduces primarily within the palm tree , Attalea butyracea . T . cruzi human infection prevalence in this area is generally between 2 . 9% and 5% , with a relatively high peridomestic and domestic vector infection rate with T . cruzi [33]–[35] . In Panama , the opportunistic mesopredator Didelphis marsupialis , the common opossum , is believed to be a key reservoir for T . cruzi infection [36] , [37] . The land surrounding protected areas of the Panama canal has undergone a high degree of deforestation in the past 40 or so years , converting contiguous old growth and late secondary growth forests to a landscape of small patches of riparian forest remnants surrounded by a sea of cattle pasture , agricultural development , and human settlement , with small patches of regenerating forest from abandoned pasture . The landscape heterogeneity of the Panama Canal thus offers unique opportunities to understand the role of species diversity in a changing landscape on vector infection patterns . Specifically , here we test the following hypotheses: 1 ) Species composition and diversity of blood meals for R . pallescens vectors should differ as a function of habitat type , 2 ) Vectors are expected to feed off a higher diversity of hosts in less disturbed habitats , where there is a presumed higher overall vertebrate diversity , 3 ) If a ‘dilution effect’ is driving vector infection prevalence rate , then host blood meal diversity should be inversely related to related to T . cruzi vector infection prevalence ( a dilution effect ) , 4 ) Vector infection prevalence should also depend upon the life histories of species upon which vectors are feeding . The proportion of infected vectors in a particular site should be correlated to the life histories of the dominant species upon which vectors feed at a particular site . As the intrinsic rate of increase ( the maximum per capita rate of population increase ) of mammalian hosts increases , as may be expected to occur in disturbed habitats , the transmission of T . cruzi and the corresponding proportion of T . cruzi infected vectors may increase , mainly due to increased rates of production of susceptible hosts , and potential immune-mediated mechanisms . It is also possible that an overall reduction in diversity may skew blood meal community composition to increase the relative proportion of ‘fast-living’ species fed upon by R . pallescens .
This study took place in protected areas ( contiguous forest sites ) and human-dominated landscapes with a low ( contiguous forest ) to high level of human disturbance to the east and west of the Panama Canal . The study area included deforested rural landscapes and contiguously forested protected areas flanking the Panama Canal and encompassed an area of over 600 km2 . This area is classified as lowland tropical moist forest [38] . Five different habitat types were sampled for R . pallescens: contiguous late secondary forest , early secondary forest fragments , mid secondary forest remnants or fragments , cattle pasture , and peridomiciliary areas . Contiguous late secondary forest sites were located in a protected national park adjacent to the Panama Canal . These sites have a known land use history and forest age ( approximately 75–100 years old ) [39] . Early secondary forest fragments were areas of abandoned pasture or cropland undergoing forest succession . These sites were approximately 30 to 50 years old , and most trees within the early secondary sites did not exceed 10 m in height . There was also a predominance of lianas in most of these early secondary forest fragments . Mid secondary forest remnants or fragments were forest patches remaining after large-scale deforestation of late secondary or mature forest . Most of these mid-secondary patches were highly disturbed , as most of the economically valuable adult trees were previously harvested from these sites , and the forest floor of most of these forest patches were heavily trampled by cattle . Peridomiciliary areas consisted of home gardens or yards located within 100 m of a human dwelling . The gardens and yards surrounding domiciles were highly variable , some with well-manicured lawns , others with tall grass or located near a forest patch , and some sites with a large number of domestic animals ( domestic fowl , dogs ) . The early secondary forest fragments , mid secondary forest remnants , cattle pasture , and peridomiciliary areas were all located on private property . Permission from the owners was obtained to sample palms for bugs at each site . The contiguous late secondary forest sites took place within Soberania National Park . Permission was obtained from park authorities to sample palms for bugs . Furthermore , for all sample sites , collecting permits for sampling of R . pallescens were obtained from the Autoridad Nacional del Ambiente ( Environmental National Authority ) of Panama . Seven replicate sites were chosen from each of the five habitat types , comprising total of 35 sites that were sampled for R . pallescens to the west and east along the western and eastern border of the Panama Canal Area [40] . Individual sites were at least 200 m apart from each other , based on an estimated maximum flight distance estimated for Rhodnius sp . , with the majority of sites located more than 1 km apart from each other [41] . Replicate sites from identical habitat types were located at least 600 m apart from one another . Palms were sampled once from each site during the wet season , between May 2007 and December 2007 , to control for possible effects of season on R . pallescens abundance . Sites from multiple habitat types were sampled within each month , and an attempt was made to spread the sampling of different habitat types evenly across the wet season . Within each site , a total of five palms were sampled; an adult Attalea butyraceae and the four nearest accessible adult A . butyracea palms . The initial palm was selected by choosing the nearest palm to a random direction and distance less than 20 m from the observer . The height at the top of the crown base , number and ripeness of fruit racimes , and presence of animal ( bird and/or mammal nests or resting sites ) were also recorded for each palm . Three mouse baited traps modified from previously described methods [42]–[44] were placed within the crown of each palm , left for twenty four hours , and checked for R . pallescens the following day . Traps were approved by the Gorgas Memorial Institute Animal Care and Use Committee in accordance with Panama's regulations for animal use . After collecting the baited traps , palm crowns were searched for 10 minutes for bugs by a skilled individual . Palm crowns were accessed with a 20 foot ladder or by climbing the palm tree with a rope and harness tree climbing technique modified for palms . R . pallescens ( N = 643 ) captured from each tree were classified according to stage , weighed , and measured . Only fourth , fifth stage nymphs , and adults were weighed and measured . Using sterile scissors , triatomines were macerated in 500 l of 0 . 01 Molar , 7 . 6 pH phosphate buffered saline ( PBS ) . The macerated triatomines were centrifuged at 15 , 000 G for 10 minutes . The pellet containing portions of exoskeleton and internal organs of R . pallescens was then resuspended with a small sterile wooden dowel and subsequently centrifuged at 400× for five minutes . The supernatant was collected and centrifuged at 15 , 000 G and the pellet containing fragments of exoskeleton was frozen at −20°C . The collected supernatant was spun for a final time at 15 , 000 G for 20 minutes . The supernatant with soluble proteins from this spin was then frozen at −20°C [33] , [37] . The precipitate was suspended in 200 l of PBS and 5 l of this suspension was evaluated microscopically for the presence of trypanosomes . The rest of this suspended precipitate was frozen at −20°C until DNA extraction was performed . DNA was extracted from this suspension using a comercial kit ( Promega , Madison , WI ) . A duplex polymerase chain reaction was performed for the detection of T . cruzi and T . rangeli using an assay targeted to the 189 base pair telomeric junction of T . cruzi and a subtelomeric region of T . rangeli developed by Chiurillo et al . ( 2003 ) [45] . The primers used for T . cruzi detection were T189Fw2 ( 5′ -CCAACGCTCCGGGAAAAC-3′ ) and Tc189Rv3 ( 5′ -GCGTCTTCTCAGTATGGACTT-3′ ) . For T . rangeli detection ( results used for other studies ) primers targeted to a conserved subtelomeric region were TrF3 ( 5′ -CCCCATACAAAACACCCTT-3 ) and TrR8 ( 5′-TGGAATGACGGTGCGGCGAC-3′ ) . PCR products ( 5 l ) were mixed with loading dye and electrophoresed on a 1 . 5% agarose gel stained with ethidium bromide and evaluated by ultraviolet light for the presence of bands of a length specific for T . cruzi ( 100 bp ) and T . rangeli ( 170 bp ) . Positive and negative controls were run for each reaction . In order to identify the vertebrate species present in insect bloodmeals , extracted DNA from triatomines ( N = 643 ) was used in a PCR assay adapted from Humair et al . 2007 that amplifies the 12S mitochondrial rRNA gene of vertebrates [46] . Due to positive template bias in the PCR reaction , it is unlikely that this assay would result in the detection of multiple blood meal sources in a single vector . The primers used to amplify the approximately 145 bp fragment of the 12S rRNA gene were 12S-6F ( 5–CAAACTGGGATTAGATACC–3 ) and 12S-9R ( 5–AGAACAGGCTCCTCTAG–3 ) . Primers were obtained from Integrated DNA Technology services , USA . A 25 l reaction was prepared for PCR amplification with 3 . 0 mM MgCl2 ( Fermentas ) , 0 . 2 mM dNTPs ( Qiagen ) , 0 . 8 M of each primer , of Taq buffer , and 2 . 35 U of Taq DNA polymerase ( Fermentas ) . Five microliters of triatomine DNA template was added to each sample . Positive and negative controls were run for each reaction . PCR reaction conditions were as follows: touchdown - initial denaturation 3 minutes at 94°C , burst cycle 20 seconds at 94°C , 30 seconds at 60°C , and 30 seconds at 72°C . Forty cycles of the following were then performed , with the annealing temperature being lowered by 1°C until reaching 52°C , : 20 seconds at 94°C , 30 seconds at 52°C , and 30 seconds at 72°C . There was a final extension step of 7 minutes at 72°C . After the reactions , 1 l of PCR product was mixed with 5 l of loading dye and run on a 2% agarose gel that was stained with ethidium bromide in order to detect if the reaction was able to amplify vertebrate DNA in the bug . PCR products were stored at −20°C until the final pre-sequencing purification step . PCR products were then purified with a high throughput adaptation of gel extraction followed by vacuum manifold PCR product purification using a QIAquick 96 PCR Purification Kit ( Qiagen , Valencia , CA 91355 ) following manufacturer's instructions . After purification , PCR products were then tested for purity and DNA concentration with a Nanodrop spectrophotometer and sent for sequencing to the University of Georgia Bioinformatics laboratory . Sequences were evaluated for quality by checking chromatogram patterns as well as double-nucleotide peaks , that may indicate blood meals from more than one host species . Sequences were identified to genus and species by performing a nucleotide BLAST ( Basic Local Alignment Search Tool ) using the NCBI Nucleotide collection ( nr/nt ) database and comparison of the unknown sequence to a known species . The cutoff for accepting a species or genus sequence was typically an 85% to a complete identity match and an E-value ( probability that the sequences align due to random chance given the sequences in the database ) less than 1×10−10 . Percent identity matches are shown in Table S1 . If the local Panamanian species of a 12SrRNA gene sequence was not available on the NCBI database , but a congeneric species not found in the study area gave an adequate sequence match , then the sample was identified to the genus level . The host species diversity of blood meals identified by molecular analysis was quantified for a respective site ( sites with only one blood meal identified were discarded from this analysis ) . For each site , the number of different mammalian blood meal species was recorded as host species richness . In order to account for the species number and relative proportion of each species blood meal identified for each site in relation to diversity , the Shannon Weiner diversity index ( H′ ) was calculated , substituting the number of different blood meal species for the number of species . To assess the degree of similarity in identified blood meals across our study sites we estimated the Horn distance [47] among all our study sites . We chose the Horn distance because it measures the faunistical similarities between two sites , while weighting differences in the abundance of different taxa . The index is 1 for a perfect similarity and 0 for a perfect mismatch . We then estimated the spatial autocorrelation of the Horn distance in our samples by performing a Mantel test of the Horn distances as a function of the geographical distance between the sites [48] . We also tested if clustering patterns on the diversity of blood meal sources were shaped by the kind of habitat where we sampled the blood-fed kissing bugs . For this purpose , we estimated the Simpson species similarity index , an index that measures faunistic overlap focusing only on patterns of taxa presence/absence [49] . The Simpson index can have values between 0 and 1 , with an interpretation similar to Horn distances . We used the Simpson index estimates from all sites to build agglomerative clusters , which graphically depict the similarity between sites by hierarchically clustering the most similar observations [50] . The rmax value ( maximum intrinsic rate of increase ) for each mammalian species fed upon by each bug ( based on blood meal identification results ) was recorded from published estimates in the literature that estimated rmax from Cole's equation [51] , [52] . When data was not available , rmax was estimated from Cole's equation using published data on age of first reproduction , annual birth rate , and lifespan ( http://www . demogr . mpg . de/longevityrecords/ ) . A mean rmax ( maximum intrinsic rate of increase ) score per site was estimated for each study site by adding together the rmax values for the blood meal species present at a site divided by the number of different mammal species identified at that site . The weighted mean , standard deviation , and kurtosis of the rmax values from each site were also calculated . For all analyses , we used RCRAN version R 2 . 7 . 1 GUI 1 . 25 ( 5166 ) . R Development Core Team ( 2008 ) . R: A language and environment for statistical computing . R Foundation for Statistical Computing , Vienna , Austria . ISBN 3-900051-07-0 , URL http://www . R-project . org . Fisher's chi test was used to measure dependence between taxonomic order and species blood meal identification and habitat type . A general linear model with binomial errors was used to predict of effects tamandua , opossum , and primate blood meal availability on vector infection prevalence with T . cruzi . Regression tree models were used to evaluate the relationship between the proportion of T . cruzi infected vectors ( response variable ) from each site ( calculated from the total number of bugs tested for T . cruzi ) , and the following independent variables: habitat type , blood meal species diversity , mammalian blood meal species richness , mean rmax value for mammal blood meal species identified at each site , as well as the standard deviation and kurtosis of rmax estimations of mammalian species that the vectors fed upon . Regression tree models are a useful way to analyze complex ecological data , including a combination of categorical and numerical data whose relationships between variables may be nonlinear or difficult to an analyze by standard statistical modeling procedures , as well as missing values [53] , [54] , as was the case with this data .
Overall , 74 . 3% , ( 478/643 ) of vectors were infected with T . cruzi . Figure 1 shows T . cruzi vector infection rate across habitat types , ranging from 58% in contiguous forests to 85% in peridomiciliary sites . The vector infection rate with T . cruzi rate was significantly higher in mid secondary forest remnants ( p<0 . 01 ) and peridomiciliary sites ( p<0 . 01 ) as compared to contiguous forests . Data showed differences in blood meal species composition as well as the relative proportion of each order present in blood meals across habitat types . Blood meals were identified in 40 . 3% ( 259/643 ) of vectors tested . Blood meal analysis identified 42 different species fed upon by R . pallescens . Tables 1 and 2 show class , genera , and species identity of blood meals . Mammals made up 88 . 6% of blood meals across all habitat types . The rest of the blood meals were composed of birds ( 6 . 9% ) , reptiles ( 3 . 1% ) , and amphibians ( 1 . 5% ) . There was a significant association between order and habitat type ( Fisher's exact test 2 , p = 0 . 002 ) and species identification and habitat type ( Fisher's exact test 2 , p = 0 . 003 ) . Table 3 shows the proportion of blood meals by taxonomic order in each habitat type . Blood meal composition differed across sites with varying degrees of anthropogenic disturbance . Xenarthrans made up the highest proportion of blood meals in all habitats ( between 44–54% of blood meals ) except for peridomiciliary areas , where marsupial blood meals were ranked first , making up 28% ( N = 14 ) of blood meals , with Xenarthrans making up 14% ( N = 7 ) of blood meals . In contiguous forest sites , tamanduas Tamandua mexicana made up 39 . 4% ( N = 13 ) and Choloepus hoffmani , two-toed sloths , comprised 60 . 6% ( N = 20 ) of Xenarthran blood meals . However , in mid-secondary , early secondary , and pasture sites , the number of tamandua blood meals decreased , making up 5 . 3% , 6 . 3% , and 7 . 0% of Xenarthran blood meals , respectively , with sloths comprising over 90% of the Xenarthran blood meals in these sites . In peridomiciliary sites , only sloths , and no tamanduas , were detected in blood meals . Additionally , primates Cebus capucinus and Allouatta palliata blood meal isolations were highest in contiguous forests , comprising 28% of blood meals in this habitat type , and between 4 . 2 to 10 . 5% of blood meals in the deforested landscape habitats ( Figure 1 ) . Domestic animal blood meals identified in peridomestic habitats included cows ( N = 3 ) , swine ( N = 2 ) , domestic dogs ( N = 1 ) , turkey ( N = 1 ) , and peacock ( N = 1 ) . Domestic animal blood meals identified in cattle pasture included cows , chicken , turkey , swine , and domestic dog . In early secondary forest fragments and mid secondary forest remnants , domestic animals that R . pallescens fed from were cow , domestic dog , and chicken . Similarity in blood meal vertebrate class ( Figure S1A ) and order ( Figure S1B ) was independent of geographical distance among the sites where we sampled blood-fed R . pallescens . Also , there was no evident clustering of the blood sources that matched the different habitats we sampled at the vertebrate class ( Figure S1C ) and order ( Figure S1D ) level . Figure 2 shows results of the regression tree analysis evaluating the relationship between host life history , host blood meal species diversity , and habitat type variables and the response value , that is the predicted T . cruzi vector infection rate at each site . Explanatory variables were mean rmax data for blood meal species per site , kurtosis and standard deviation of rmax value of blood meals per site , habitat type ( type = Contig-contiguous forest , Past-pasture , ES-early secondary forest fragment , MS-mid secondary forest remnant , PD-peridomicilary ) of each site , and blood meal mammalian host species diversity ( calculated as Shannon-Weiner diversity index for mammalian blood meals each site ) . Each tree split leads to a non-terminal ( surrounded by a circle ) or terminal ( surounded by a rectangle ) node . Each of four splits is labeled with a particular variable and values that determined the split . The main split at the top of the tree shows the predictor variable responsible for the largest variance change in the explanatory variable ( in this case mean rmax value of host blood meals at from each site ) , and the total number of sites ( n = 32 ) evaluated for this tree . Each terminal and non-terminal node is labeled with the predicted infection prevalence rate and the number of sites that corresponded to the particular node . The predicted infection prevalence rate for a site is shown at each of the six terminal nodes . Based on this regression tree , mean rmax score and the statistical distribution of the rmax score ( kurtosis and standard deviation ) , species diversity , and habitat type are key factors influencing vector infection prevalence . The tree explained 68% of the total variance in the response variable ( vector infection rate ) . Predicted vector infection rate is lowest for contiguous and pasture sites containing a mean blood meal species rmax of under 0 . 35 and highest for sites with a rmax mean greater than 0 . 35 , a kurtosis of over 1 . 5 , and a standard deviation of over 1 . 6 . Higher species diversity ( Shannon-Weiner index greater than or equal to 0 . 51 ) , was associated with lower vector infection rates . Table 4 shows that the presence of opossums ( D . marsupialis and Metachirus nudicaudatus in blood meals is also positively associated with vector infection rate within a particular site .
There is a relatively high R . pallescens vector infection prevalence ( 80–90% ) with T . cruzi in this region of Panama [33] , [37] and northern Costa Rica ( 100% prevalence ) , [55] , compared to reports of R . pallescens infection in eastern Panama ( 17 . 8% ) [56] . With the exception of Rhodnius spp . in Attalea palms in Brazil , with a vector infection prevalence between 41% to 47% [57] , T . cruzi vector infection prevalence in Rhodnius spp . in Central and South America ranges from 1 . 9% to 19 . 1% [58]–[66] . Predominant feeding from mammalian hosts , the only competent reservoirs for T . cruzi infection , may be an important explanation for relatively high R . pallescens vector infection indices in Panama and in those reported by Teixeira et al . ( 2001 ) [57] . The palm Attalea butyracea , the main habitat for R . pallescens , likely provides a key nesting space for mammals , as well as birds , reptiles , and amphibians [55] , [57] . Furthermore , in deforested areas or within forest remnants , palms may provide refuges for vertebrates , particularly mammals , in sites where other hiding or nesting sites have been disturbed , and they may use these palms more frequently than in more undisturbed habitats . Although there was a relatively high number of species identified in blood meals , these results likely underestimate actual blood meal species diversity in each site because the molecular test used may not be able to distinguish between some closely related species . The vast majority of blood meals ( 88 . 6% ) were identified from mammals ( competent hosts ) as compared to birds , reptiles , and amphibians that cannot transmit T . cruzi nor T . rangeli . In this study , the species composition of R . pallescens blood meals differed across habitat types . In Panama , Pineda et al . ( 2008 ) encountered a predominance of mammal blood meals , particularly wild mammals , from peridomestic and domestic sites [37] . In early secondary fragments , mid secondary forest remnants , and peridomiciliary sites , the blood meal species richness was higher than in contiguous forests . Domestic animal blood meals were only detected in disturbed habitats ( forest fragments , cattle pasture , and peridomiciliary areas ) . As an order , Xenarthrans ( sloths and tamanduas ) comprised the majority of R . pallescens blood meals identified across all habitats . The pattern where kissing bugs feed on whatever vertebrate is present in a given location has been observed in many species , and may be related with the potential of R . pallescens to effectively adapt to disturbed landscapes [67] . Overall , sloths ( Choloepus hoffmanni ) dominated blood meals in all habitat types . Sloths may be an attractive blood meal for R . pallescens , and good hosts for trypanosomes and bug populations . In addition to habits of resting in palm crowns , the sloth's slow metabolism and relatively slow movements may prevent them from rapidly removing feeding bugs by grooming or scratching , allowing bugs to feed , defecate on the host , and assist in maintenance of T . cruzi transmission . Although most blood meals identified were arboreal or scansorial species , a few terrestrial species , such as dog , pig , and cow , were identified . Although most terrestrial species blood meals were identified from adult bugs , terrestrial mammal blood meals were also identified in nymphs , suggesting that nymphs may descend to the ground near palm trees to feed . Alternatively , nymphs may secondarily feed from engorged adults ( a phenomenon known as ‘kleptohemodeipnonism’ ) [68] who fed from terrestrial species and returned to palm trees to rest , and become infected by them [69] . Host composition may also play an important role in driving infection patterns in landscapes [70] . In this study , host communities change across habitats , with a marked increase in opossum blood meals in peridomestic sites . Sloths remain the top ranking blood meal across most habitat types , with the exception of peridomiciliary areas , where marsupials ( Didelphis and Metachirus ) dominate . Because marsupials are believed to be a particularly competent reservoir for T . cruzi infections [35]–[37] , [56] , [71] , they may play an important role in driving the T . cruzi vector infection prevalence up in peridomiciliary sites . Results from this study suggest that important factors determining T . cruzi vector infection rate in R . pallescens include mammal species composition , life history strategies of mammalian hosts that are fed upon , blood meal species diversity , and habitat type . There is a significantly positive association between the proportion of blood meals composed of opossums and vector infection rate ( Table 4 ) . This is not suprising , because opossums are believed to be important reservoirs of T . cruzi [23] , [36] , [56] , [57] , [72] , [73] . According to regression tree analyses , the mean rmax value for mammals fed upon by bugs at a particular site was a key factor in determing vector infection rate , with higher mean rmax values at each site tending towards a higher vector infection rate . At values of rmax greater than 0 . 35 , the statistical distribution ( kurtosis and standard deviation ) of rmax values of mammal species that R . pallescens fed upon was also important . Large , relatively long-lived species ( e . g . primates ) , with low rmax values , may not be expected to be as important to long term T . cruzi transmission as compared to a shorter lived species with a higher intrinsic rate of increase . Long lived species may develop long-lasting acquired immunity to trypanosome infection , decreasing the probability of being a source of vector-borne transmission to susceptible individuals . Typically , circulating parasitemias after reinfection with T . cruzi after the course of initial infection are reduced as compared to the initial infection due to acquired immunity 74–76 . However , in short-lived , relatively smaller sized individuals such as the opossum , infective adults sharing a nest with juveniles may transmit the disease rapidly to vectors , which can then transmit the parasite to susceptible offspring , helping maintain T . cruzi infections in bug populations [77] . There is also the possibility of direct transmission via anal glands of opossums [71] . High kurtosis and relatively high standard deviation of rmax values associated with high site-level vector infection indices suggests that a few key mammal species may contribute disproportionally to vector infection . Alternatively , it is possible that species with higher rmax values , such as opossums , have a higher tolerance to trypanosome infection , making them particularly competent disease reservoirs . For example , Didelphis are commonly coinfected by many types of protozoan parasites , such as Sarcocystis and Besnoitia [78] , and may be able to tolerate and transmit protozoan infection with greater facility than other mammal hosts . A greater understanding of the relative susceptibility to and competence for T . cruzi infection in different Neotropical mammal species is critical to predicting trypanosome infection dynamics and disease risk across the Neotropics . The reason why pastures and contiguous forests have lower vector infection rates may be due to a combination of harboring host species with lower intrinsic rates of population increase . Furthermore , mammal reservoir hosts may not nest or spend a long time resting in palm crowns in cattle pasture due to increased exposure to sun and rain , and prefer to nest in trees in relatively sheltered sites such as early and mid secondary forest fragments , and peridomestic areas , which tend to be surrounded by other trees or a more complex vegetation structure . However , in sites with low mean intrinsic rates of reproduction of mammalian hosts ( ) , and a low diversity index ) , the predicted vector infection rate is relatively high , suggesting a dilution effect may occur under conditions where most mammalian hosts fed upon in a site are long-lived . Limitations of this study include its duration and the specificity and sensitivity of detection of the blood meal identification method . Because it was a cross sectional study , bug samples were taken only during the wet season , and transmission dynamics may change as a function of seasonality and long term environmental drivers ( e . g . climate change ) . Our method was able to identify vertebrate blood meals from approximately 40% of bugs , many of whom were thin and had not fed recently . Unfortunately , there is no ‘gold standard’ methodology for triatomine blood meal detection . Our method , while successful in amplifying small fragments of DNA from vertebrate blood meals , lacked specificity for discrimination between some mammal species and identification of particular species . For example , sequences of the 12S rRNA gene amplified for D . marsupialis and M . nudicaudatus were very similar , thus there may be error in discrimination between these species . If the 12S rRNA gene sequence for a particular species present in a blood meal was not present in the NCBI database , the BLAST search may not have been able to align a sequence with the appropriate host . Furthermore , we were unable to detect dual blood meals within an individual bug , a concern because the bugs may feed from multiple hosts . Successful development of an assay such as the reverse line blot hybridization assay used to identify blood meals in ticks [46] , [79] , or potentially using next generation sequencing would be useful in order to identify dual blood meals within kissing bugs . Additionally , the relative reservoir competence for different mammalian hosts identified is undefined and requires future host-focused studies ( xenodiagnostics , experimental infection studies ) . In summary , reservoir host life history , diversity of competent blood meal species , as well as habitat type contribute to T . cruzi vector infection in R . pallescens . Results suggest that vector infection prevalence increases with reservoir host intrinsic rate of increase , lower mammalian host diversity , and deforestation/forest fragmentation . | Understanding how host species influence vector-borne pathogen transmission in anthropogenically disturbed landscapes is important to predicting and preventing disease transmission . This study evaluates how host diversity , anthropogenic land use change , and host life history influence vector- borne multihost pathogen transmission in Panama , where the triatomine bug Rhodnius pallescens is the principal vector of Trypanosoma cruzi , agent of Chagas disease . We hypothesize that blood meal species composition and vector infection differ as a function of habitat disturbance , and that the host species intrinsic rate of increase is positively associated with T . cruzi vector infection . We collected R . pallescens across a gradient of anthropogenic disturbance . Blood meal species composition and T . cruzi vector infection were determined by molecular methods . Vector infection rates were higher in deforested habitats and forest fragments as compared to contiguous forests . Vectors fed primarily on mammals , likely accounting for a relatively high vector infection prevalence , and that host blood meal species composition varied across habitat types . Regression tree analysis demonstrates that higher T . cruzi vector infection indices we associated with sites that had blood meal species with higher , more variable , and more skewed rmax ( intrinsic rates of increase ) values , lower blood meal species diversity , and disturbed habitats , namely fragmented forests and peridomiciliary sites . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"evolutionary",
"ecology",
"microbiology",
"parasitic",
"diseases",
"neglected",
"tropical",
"diseases",
"veterinary",
"science",
"infectious",
"diseases",
"veterinary",
"diseases",
"zoonotic",
"diseases",
"biology",
"vectors",
"and",
"hosts",
"community",
"ec... | 2012 | Host Life History Strategy, Species Diversity, and Habitat Influence Trypanosoma cruzi Vector Infection in Changing Landscapes |
Nanoparticles introduced in living cells are capable of strongly promoting the aggregation of peptides and proteins . We use here molecular dynamics simulations to characterise in detail the process by which nanoparticle surfaces catalyse the self-assembly of peptides into fibrillar structures . The simulation of a system of hundreds of peptides over the millisecond timescale enables us to show that the mechanism of aggregation involves a first phase in which small structurally disordered oligomers assemble onto the nanoparticle and a second phase in which they evolve into highly ordered as their size increases .
With the advent of nanoscience much interest has arisen about the ways in which nanoparticles interact with biological systems , because of their potential applications in nanotechnology and effects on human health [1]–[7] . When nanoparticles are introduced in a living organism they may interact with a variety of different cellular components with yet largely unknown pathological consequences . These concerns have been articulated particularly in the case of misfolding disorders with increasing evidence , for example , about an association between exposure to heavy metals and an enhanced risk of developing Parkinson's disease [8] . Such misfolding diseases are caused by the aberrant association of peptides and proteins [9] , which result in fibrillar aggregates that share a common structure of intertwined layers of [9] . Although is well known that such aggregates are formed in a nucleation-dependent manner [9] , [10] and that very often nucleation phenomena are known to be triggered by external factors [11] , experimental reports on protein aggregation in heterogeneous systems have only begun to emerge [12]–[15] . These studies are important , since peptides and proteins in vivo often interact with a variety of potential seeding agents such as macromolecular complexes or membranes , which may strongly influence their aggregation behaviour . Indeed , it is well known that colloids [12] , [14] , [15] , lipid bilayers [16] , and liquid-air , liquid-solid or liquid-liquid interfaces [17] , [18] can have significant effects in promoting amyloid formation . It has also been recently shown that , in vivo , nanoparticles are often covered by peptides and proteins that determine their behaviour in the cell [13] , [15] . Despite these observations , the detailed processes underlying the association of proteins on surfaces or nanoparticles have so far remained elusive . In this work we use molecular dynamics simulations to investigate the molecular mechanism of peptide self-assembly in the presence of spherical nanoparticles . Although computational studies using full atomistic models have provided considerable insight into the role of fundamental forces in promoting the self-assembly of polypeptide chains , they are restricted to relatively small systems of peptides and short timescales [19]–[29] . Coarse-grained models have proven capable of following the evolution of systems composed of larger numbers of peptides over longer timescales . The most tractable models are confined to a lattice [30]–[32] , although in these cases the structural details used to represent polypeptide chain conformations are necessarily limited . Off-lattice protein models used to simulate protein aggregation include two-state models in which the protein can adopt , in addition to a native state , a state that is prone to formation [33] , [34] , two-bead models , in which each amino acid is represented by two spheres with a knowledge-based potential [35] , and fine-grained models with explicit representation of the side chains in combination with a phenomenological force field [36]–[39] . The more detailed is the protein model , the higher is the computational cost and the larger is the number of parameters required to specify the force field [40] . The studies mentioned above have investigated the process of protein self-assembly in homogeneous systems in which external factors such as nanoparticles or other molecules are absent . Only very recently , Friedman et al . investigated the process of assembly of amphiphatic peptides in the presence of lipid vescicles [41] . In the present work , we adopted an off-lattice protein model [42]–[45] , in which the protein backbone is represented as tube that embeds a chain of atoms subject to interactions that are common to all polypeptide chains , including excluded volume constraints , hydrophobic attractions , bending rigidity , and cooperative hydrogen bonds ( see Materials and Methods ) . The major strength of the model is its ability to reproduce rather accurately secondary structure elements through the excluded-volume effects due to the tube geometry [42]–[45] , which enables the use of a relatively simple force field and thus is very efficient computationally . By using this type of model , we already provided insight into the early stages of the aggregation process , to establish the existence of a general condensation-ordering transition for protein aggregation [46] , [47] , and to reveal a self-templated nucleation mechanism [48] that is able to explain a key feature observed in protein aggregation - the coupling between the initial formation of oligomeric assemblies and their subsequent rearrangement into a highly ordered structures . In this work , we show the feasibility of simulating hundreds of peptides over several milliseconds , and we characterise in detail the molecular mechanism of self-assembly of the peptides at the surface of nanoparticles . This process takes place in two steps - at first the peptides associate on the surface thus increasing their local concentration and subsequently they undergo a process of reordering into sheet structures , which is driven by the tendency to form hydrogen bonds .
We have characterised the process of nanoparticle-catalysed peptide aggregation in terms of a condensation-ordering mechanism and investigated its dependence on the nanoparticle diameter and the strength of the nanoparticle-peptide interactions . A similar mechanism of aggregation has already been observed in the absence of catalysing factors [46]–[48] , [52] , suggesting that the process of aggregation is driven in both cases by the intrinsic tendency of polypeptide chains to associate by forming ordered networks of hydrogen bonds [53] , [54] . In the case that we have studied here , the initial condensation of peptides is initiated by nanoparticle surfaces to form small disordered oligomeric structures that subsequently re-order into as their size increases . Although this mechanism will be modulated by specific sequence-dependent interactions for more complex amino acid sequences , our findings are consistent with recent experiments on seeded fibrillation [12] . These results therefore suggest that the process of protein aggregation can be speeded up by the presence of factors capable of increasing the local concentration of proteins and thus promoting the formation of disorder oligomeric assemblies whose presence in turn facilitates the conversion of soluble proteins into highly ordered fibrillar structures .
We used a modified version of the tube model [42] . In this model , residues are represented by their atoms , which are connected into a chain with a distance of 3 . 8±0 . 2 Å between neighbouring atoms . The tube geometry is approximated by assigning a diameter of 3 . 8 Å to the atoms . Neighbouring atoms are not allowed to interpenetrate . Bond angles are restricted between 82° to 148° , and an analogue of bending rigidity is introduced by means of an energetic penalty , , for values of bond angles lower than 107 . 15°; these are the same criteria used previously [42] . The introduction of is useful to mimic the constraints placed on local conformations by the presence of side chains , as usually visualised by Ramachandran plot . Hydrophobicity enters through a pairwise-additive interaction energy of ( positive or negative ) between any pair of residues and that approach closer than 7 . 5 Å . The quasi-cylindrical symmetry of the tube is broken by the geometric requirements of hydrogen bonds . These geometrical requirements were deduced from an analysis of 500 high resolution PDB native structures [55] , from which we computed the normalised histograms of distances between atoms involved in backbone-backbone hydrogen bonds which are shown in Fig . 4 . The distances we used to define the hydrogen bonds at the atom level are summarised in Table 1 . Our definitions distinguish between hydrogen bonds that belong to a helix , parallel or anti-parallel sheets . We emphasise the fact that there is not a full correspondence with the real hydrogen bonds formed between amide and carboxyl backbone groups . For instance , there are two different kinds of residue pairs facing each other in nearby anti-parallel . In the first kind , the two hydrogen bonds are formed between the two residues , whereas in the second kind , no hydrogen bond is formed between them . The two kinds alternate along the pair of nearby strands . In our definition of hydrogen bonds based on atoms , we will say that for both kind of pairs one hydrogen bond is formed between the two . Yet , we keep track of the peculiar geometry of hydrogen bonds within anti-parallel by using two different sets of distances , which we call anti-parallel 1 and anti-parallel 2 , as the distances between consecutive pairs facing each other on nearby do indeed alternate . Furthermore , we request that one residue cannot form more than two hydrogen bonds , and that the first and last atoms of a peptide do not at all . Hydrogen bonds may form cooperatively between residues and [or and for anti-parallel hydrogen bonds] , thereby gaining an additional energy of . The distance criteria for cooperative hydrogen bonds within are obtained from Fig . 4F and summarised in Table 1 . The energy of hydrogen bonds was set to , where is the thermal energy at room temperature and is the Boltzmann constant . This energy correspond to the experimental one ( 1 . 5 kCal/mol at room temperature [56] ) . Values of the hydrophobicity and stiffness parameters , and , are given in units of and the reduced temperature is . In all our simulations we set and . The ratio of a hydrogen bonding energy to hydrophobic energy is . As the number of hydrophobic contacts within an oligomer is usually about one order of magnitude larger than the number of hydrogen bonds , our choice ensures that these interactions provide similar contributions to the potential energy of the oligomer [47] . For this set of model parameter the peptide folds into a native state below the folding temperature . is the parameter which determines the strength of the interaction energy between atoms representing the peptide molecules and the seed particle . The range of the peptide seed interaction is set to 10 Å from the nanoparticle surface . We performed discontinuous molecular dynamics ( DMD ) simulations [57] , which is a fast alternative to standard molecular dynamics simulations . The main difference is that in DMD simulations the system evolves on a collision by collision basis , and requires the calculation of the collision dynamics and the search for the next collision . In the simulations we used a cubic box , of side 633 Å , and applied periodic boundary conditions . The implementation of our definition for the hydrogen bonding requires some additional consideration . In order to prevent that one residue forms three hydrogen bonds we treat the associated collision as fully elastic . In order to implement and consider cooperative hydrogen bonding we keep and update a list of all hydrogen bonds formed in the system at all times . Note that a recalculation of the hydrogen bonds formed in the system without considering this list can lead to a different result . Independent starting configurations were generated at and rapidly cooled down to at the beginning of each simulation run . We performed all our simulation in the NVT ensemble using an Anderson thermostat . In order to associate the number of collision steps performed in our simulation to a real time we measured the long time self-diffusion coefficient of our model peptide , , and matched it to experimental data . We took from the literature the value for the self-diffusion coefficient , , which was measured for lysozyme [43] . The Einstein relation for the diffusion coefficient together with the Stokes law yield where is the Boltzmann constant , is the radius of the diffusing object , and is the viscosity . The latter can be evaluated through kinetic theory as , where is the density of the viscous medium in which diffusion takes place and is the mean flight time between collision with solvent molecules setting the time scale [15] . The resulting expression for the diffusion coefficient allows us to get picoseconds as an estimate of the real time corresponding to one collision step in our molecular dynamics simulations . We use as an estimate of for lysozyme , whereas we take as the average radius of gyration of the peptide as found in our simulations . Hence , the total number of collision steps , 4×109 , performed in every simulation corresponds qualitatively to 0 . 78 milliseconds . | Protein misfolding and aggregation are associated with a wide variety of human disorders , which include Alzheimer's and Parkinson's diseases and late onset diabetes . It has been recently realised that the process of aggregation may be triggered by the presence of nanoparticles . We use here molecular dynamics simulations to characterise the molecular mechanism by which such nanoparticles are capable of enhancing the rate of formation of peptide aggregates . Our findings indicate that nanoparticle surfaces act as a catalyst that increases the local concentration of peptides , thus facilitating their subsequent assembly into stable fibrillar structures . The approach that we present , in addition to providing a description of the process of aggregation of peptides in the presence of nanoparticles , will enable the study of the mechanism of action of a variety of other potential aggregation-promoting agents present in living organisms , including lipid membranes and other cellular components . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biophysics",
"computational",
"biology"
] | 2009 | A Condensation-Ordering Mechanism in Nanoparticle-Catalyzed Peptide Aggregation |
Rift Valley fever virus ( RVFV ) belongs to the family of Phenuiviridae within the order of Bunyavirales . The virus may cause fatal disease both in livestock and humans , and therefore , is of great economical and public health relevance . In analogy to the influenza virus polymerase complex , the bunyavirus L protein is assumed to bind to and cleave off cap structures of cellular mRNAs to prime viral transcription . However , even though the presence of an endonuclease in the N-terminal domain of the L protein has been demonstrated for several bunyaviruses , there is no evidence for a cap-binding site within the L protein . We solved the structure of a C-terminal 117 amino acid-long domain of the RVFV L protein by X-ray crystallography . The overall fold of the domain shows high similarity to influenza virus PB2 cap-binding domain and the putative non-functional cap-binding domain of reptarenaviruses . Upon co-crystallization with m7GTP , we detected the cap-analogue bound between two aromatic side chains as it has been described for other cap-binding proteins . We observed weak but specific interaction with m7GTP rather than GTP in vitro using isothermal titration calorimetry . The importance of m7GTP-binding residues for viral transcription was validated using a RVFV minigenome system . In summary , we provide structural and functional evidence for a cap-binding site located within the L protein of a virus from the Bunyavirales order .
Rift Valley fever virus ( RVFV ) belongs to the family of Phenuiviridae within the order of Bunyavirales with a single stranded RNA genome in negative orientation ( https://talk . ictvonline . org/taxonomy/ ) . RVFV is endemic to sub-Saharan African countries but has also spread to the Arabian Peninsula . The virus infects ruminants and pseudoruminants leading to abortions in pregnant animals and high mortality among young animals . The virus can also be transmitted to humans causing febrile illness with the possibility for severe disease even with fatal outcome [1 , 2] . Due to the high economical burden of RVFV outbreaks among livestock , the possibility of severe human disease without effective antiviral treatment options and the epidemic potential RVFV is listed on the WHO R&D Blueprint . The WHO urges to focus R&D on this pathogen to develop medical countermeasures [3] . All viruses from the order of Bunyavirales amplify in the cell cytoplasm and the key factor for viral replication is the large L protein ( 220–450 kDa ) . The L protein is a multifunctional enzyme catalyzing genome replication as well as viral transcription . Whereas genome replication is initiated de novo by a prime-and-realign mechanism [4–7] , viral transcription is a primer-dependent process . Messenger RNAs of most segmented negative sense RNA viruses , including bunyaviruses , contain additional non-templated host-derived sequences at their 5' ends . Therefore , cap-snatching was proposed to be a common mechanism in these viruses to initiate viral transcription [5 , 6 , 8–10] . As demonstrated for the influenza virus polymerase complex [11] , the bunyavirus L protein is assumed to bind to cap structures of cellular mRNAs , while an endonuclease cleaves the mRNA some nucleotides downstream of the cap creating a short RNA fragment . This capped RNA fragment is subsequently used to prime viral transcription . The length of these sequences differs between the virus families [6 , 8–10 , 12–15] . This hypothesis is strengthened by the identification and biochemical characterization of an endonuclease in the N-terminus of the L protein for several viruses of the Bunyavirales order [7 , 16–21] . However , there is no convincing evidence for a cap-binding site within the L protein so far . Mutational studies on the Lassa virus and RVFV L proteins using minireplicon systems revealed a specific role of the C-terminus in viral transcription [22 , 23] . A crystal structure of a C-terminal L protein fragment from a reptarenavirus showed high structural similarity to influenza virus PB2 cap-binding domain but the reptarenavirus domain was lacking the expected cap-binding site arrangement of two aromatic side chains as well as any in vitro cap-binding activity [21] . The structure of the La Crosse virus L protein , published in 2015 by Gerlach et al . [24] , is missing the C-terminal region , allowing only for the conclusion that there is no cap-binding domain present in any other part of the L protein . We solved the structure of a C-terminal domain of the RVFV L protein and demonstrated specific binding to a cap-structure . We validated the importance of several specific residues implicated in cap-binding for viral transcription using a RVFV minigenome system . Furthermore , we compared the new structure with the structures of influenza virus cap-binding domain as well as reptarenavirus putative cap-binding domain and observed a high degree of similarity between the structures but also striking differences in the binding site composition and binding mode of the cap . In summary , we present evidence for a cap-binding site located within the L protein of a virus from the Bunyavirales order . This work also provides the basis for the design of specific compounds targeting RVFV cap-binding domain .
A secondary structure based sequence alignment of the C-termini of phlebo- and closely related banyangvirus L proteins was created to identify the region of the L protein , which might contain the potential cap-binding site ( S1 Fig ) . Guided by the information about the structural composition of cap-binding domains from influenza virus PB2 and the putative cap-binding domain of California Academy of Sciences virus ( CASV ) L protein , we searched for several β-strands interspersed with one to three α-helices . The region between residues 1677 and 2008 of the RVFV L protein was identified as target region . We designed and cloned six different constructs from this area ( putative cap-binding domain CBD 1–6 ) and tested for soluble expression in Escherichia coli . Construct details and expression data are summarized in S2 Fig . Two of the proteins could be solubly expressed ( CBD2: residues 1677–1827; and CBD4: residues 1706–1827 ) but not well purified . Therefore , we designed further constructs ( CBD 7–14 ) based on these soluble candidates and obtained two proteins suitable for crystallization experiments: CBD9 ( residues 1719–1827 ) and CBD13 ( residues 1706–1822 ) . Both CBD9 and CBD13 proteins were crystallized . During the purification process up to 6 mM of m7GTP was added , as it significantly reduced precipitation of the protein at low concentrations . Only CBD13 crystals were of good quality . This protein crystallized in space group P212121 with two molecules per asymmetric unit and the structure could be refined to a resolution of 1 . 5 Å ( Fig 1A , S1 Table , PDB ID 6QHG ) . Thus , from now on CBD13 will be referred to as RVFV CBD . Except for the N-terminal residues 1706–1707 in chain B clear electron density was observed for the whole structure . RVFV CBD crystallized as a dimer ( Fig 1A ) , which is not fully symmetric ( Fig 1B ) , with a buried surface area of ~430 Å2 between the monomers ( see also S3 Fig ) . The most prominent difference between the monomers is observed between residues 1718 and 1729 , which constitute a β-hairpin . This β-hairpin has a slightly different position in both chains ( Fig 1B , β-hairpin colored in teal ) . In solution , the protein appears to be monomeric , as revealed by size-exclusion chromatography ( S4 Fig , buffer composition: 50 mM Na-phosphate , pH 6 . 5 , 150 mM NaCl , 10% ( w/v ) glycerol ) . The structure of RVFV CBD consists of a β-sheet formed by seven β-strands , an additional β-hairpin as well as a long α-helix ( Fig 1A and 1B ) . Overall the structure seems to be very rigid with low B-factors and only the β-hairpin shows some flexibility ( Fig 1C ) . During the purification optimization process , it was discovered that concentration of the protein to more than 4 . 5 mg/ml was only possible after addition of m7GTP , a cap-analogue . Thus , m7GTP was present in all crystallization setups and additional electron density corresponding to an m7GTP ligand was indeed clearly visible in both chains of the RVFV CBD structure ( Fig 2A , S5A Fig ) . The m7GTP molecule was bound in a small , mainly hydrophobic pocket between two aromatic amino acid side chains , F1713 and Y1728 ( Fig 2 ) . Y1728 is located in the β-hairpin ( βH2 ) and F1713 in the first β-strand ( β1 ) . The phosphate moiety of the cap structure interacts with R1716 side chain , which is extending from the loop between the first β-strand and the β-hairpin . Further contacts with the m7GTP involve M1782 , interacting with the guanine moiety and ribose of the cap-structure , as well as Q1717 , interacting with the guanine moiety ( Fig 2B and 2C , S5B Fig , S2 Table ) . F1713 , Y1728 and Q1717 are chemically conserved among phlebo- and banyangviruses , whereas for M1782 and R1716 the degree of conservation within the chemical class is about 95% and 20% , respectively ( Fig 3 ) . For chain A , some residues from neighboring molecules in the crystal also interact with the tri-phosphate of m7GTP . Due to the crystal symmetry , this is not the case for the m7GTP of chain B and the electron density for the ligand in chain B is less clear than in chain A ( compare Fig 2A and S5A Fig ) . In summary , RVFV CBD was crystallized with an m7GTP ligand bound between two aromatic side chains . Such an arrangement is found in several cellular and viral cap-binding proteins , as reviewed by Fechter and Brownlee in 2005 [25] . To verify that this binding pocket is specific for a methylated nucleotide , such as a cap-structure , we performed thermal stability assays using influenza virus PB2 cap-binding domain as a control . Addition of m7GTP to influenza virus PB2 cap-binding domain stabilized the protein , resulting in a higher melting temperature ( ΔTm = 9°C in the presence of 10 mM m7GTP ) . We tested thermal stability of RVFV CBD in the presence of m7GTP , GTP , ATP and m7GpppG . We observed a specific stabilization of RVFV CBD by m7GTP as well as m7GpppG ( ΔTm = 3–4°C ) compared to no or even negative effects on protein stability after addition of GTP or ATP , respectively ( Fig 4A ) . Isothermal titration calorimetry ( ITC ) measurements revealed a KD of ~737 μM of RVFV CBD for m7GTP ( Fig 4B ) , which is considerably higher than the KD of 1 . 5 μM reported for m7GTP binding to influenza virus PB2 [26] . For GTP the KD is too high to be determined by ITC under the same conditions as used for m7GTP ( Fig 4B ) . Thus , ITC experiments confirmed specificity of RVFV CBD for m7GTP and not GTP . In summary , we could demonstrate a clear binding preference of RVFV CBD to m7GTP over GTP or ATP in vitro . The RVFV minireplicon system [23] was used to test the effect of mutations in the cap-binding site on the ability of the L protein to transcribe and replicate the viral genome . By using an ambisense minigenome we were able to discriminate between ( 1 ) mRNA production , a process depending on the cap-snatching mechanism for priming , and ( 2 ) antigenome synthesis , which is independent of a primer . Synthesis of these two RNA species was measured via expression of renilla luciferase ( ( Ren-Luc ) reflecting essentially viral transcription ) and quantitative analysis of bands of antigenomic RNA and renilla luciferase mRNA on a northern blot . We know from previous experiments with the Lassa virus replicon system [22 , 27 , 28] that modification of L protein residues often affects the RNA polymerase function globally , which manifests as partial or complete defect in synthesis of all viral RNA species , i . e . replicative intermediates ( antigenome ) and mRNA . However , residues selectively important for viral transcription including cap-binding are expected to show a specific phenotype characterized by ( i ) reduced Ren-Luc signals indicating reduced mRNA synthesis , ( ii ) wild-type-like antigenome levels in the northern blot indicating functional RNA polymerase and ( iii ) a reduced mRNA-to-antigenome ratio as calculated from the quantitative northern blot data indicating a specific defect in transcription vs . replication ( S3 Table ) . We exchanged eight residues of the cap-binding pocket and surrounding areas both for chemically similar and different amino acids ( Fig 5 , S3 Table ) . Aromatic residues F1713 and Y1728 , relevant for stacking interaction with the m7GTP in the structure ( Fig 2B and 2C ) , could be changed to other aromatic amino acids ( F , Y , W ) without loss in transcriptional activity , even though they are almost completely conserved among phlebo- and banyangviruses ( Fig 3 ) . Mutation of those aromatic residues to chemically different amino acids resulted either in a complete loss of L protein activity ( F1713 ) or a selective defect in viral transcription ( Y1728 , Fig 5 , columns of mutants with selective transcription defect are colored in red ) . Substitution of R1716 by chemically different amino acids also resulted in a selective transcription defect , which is somehow surprising considering the low degree of conservation among phlebo- and banyangviruses ( Fig 3 ) . A selective defect in viral transcription was also caused by mutations of residues Q1717 and N1749 , although this effect was not observed for all substitutions . Residue M1782 could not be exchanged for other amino acids without a strong negative effect on general L protein function . We also included residue W1778 in our mutational study , because this is the only fairly conserved aromatic residue located in the α-helix , although it is not part of the cap-binding site ( Fig 2C ) . W1778 is obviously of importance for general L protein function as mutations to non-aromatic amino acids resulted in a complete loss of L protein activity , whereas the exchange for a phenylalanine had no effect . In summary , changes of the chemical properties of side chains directly interacting with m7GTP in the RVFV CBD structure mostly result in selective transcription defects in the context of the full-length L protein . These results support our structural data and underline the importance of the identified m7GTP-binding pocket for viral transcription . The overall structure of RVFV CBD is similar to the known structures of influenza A virus PB2 cap-binding domain [29] and CASV putative cap-binding domain [21] ( Fig 6B ) . Topology diagrams in Fig 6C provide an overview of the structural compositions of these proteins . They all consist of a large β-sheet packed against a long α-helix ( α1 ) . The β-hairpin present in the PB2 and RVFV CBD structures ( βH1+2 ) is replaced by a long loop in the CASV structure . This β-hairpin is much longer in the PB2 structure compared to the RVFV structure . The large β-sheet is very twisted in PB2 and consists of 8 β-strands whereas it is quite straight and only composed of 5 β-strands in the CASV structure . RVFV CBD contains a moderately twisted 7-stranded β-sheet and is thus structurally an intermediate between CASV and PB2 structures . The PB2 cap-binding domain is larger than CASV putative cap-binding domain and RVFV CBD . The structural similarity between those three protein domains is striking , considering the low degree of sequence conservation ( S6 and S7 Figs ) . In many cap-binding proteins like eukaryotic initiation factor 4E ( eIF4E ) , cellular cap-binding complex ( CBC ) or influenza virus PB2 , the cap structures are bound between two aromatic side chains [25 , 29] . This is also true for RVFV CBD . However , in the PB2 structure the two residues that bind the guanine moiety of the cap are located in the β-hairpin ( H357 in βH2 ) and at the end of the long α-helix ( F404 in α1 ) , respectively ( Fig 6B , right panel ) . Additionally , the ribose interacts with a phenylalanine side chain ( F323 in β1 ) extending from the first β-strand of the large β-sheet ( Fig 6B , right panel , S5C Fig ) . In RVFV CBD , no aromatic amino acid is present at the end of the long α-helix ( α1 ) . Instead , the guanine moiety is bound between two aromatic side chains extending from the β-hairpin ( Y1728 in βH2 ) and the first β-strand ( F1713 in β1 ) ( Fig 6B , middle panel , Fig 2C , S5B Fig ) , which are also highly conserved among phlebo- and banyangviruses ( Fig 3 ) . The latter residue , F1713 , corresponds to F323 of influenza virus PB2 , which in PB2 interacts with the ribose . In RVFV CBD methionine 1782 at the end of the long α-helix ( α1 ) takes the role of positioning the ribose ( Fig 2C , S8 Fig ) . In case of the CASV C-terminal domain , there is a conserved aromatic residue ( Y1872 in α1 ) located at the end of the long α-helix , similar to F404 in influenza virus PB2 , but no second residue in a conformation and distance suitable for a stacking interaction with a cap-structure is present ( Fig 6B , left panel ) . Among phlebo- and banyangviruses , we found only one aromatic residue near the end of the long α-helix ( α1 ) , W1818 , which is conserved in about 85% of the analyzed sequences ( Fig 3 ) . The tryptophan side chain , however , would be unable to interact with m7GTP in our structure as it is covered by M1782 ( Fig 2C ) . In summary , although the tertiary structure of the ( putative ) cap-binding domains of influenza virus , CASV , and RVFV is highly conserved , the atomic details of the cap binding sites differ significantly .
Cap-snatching was first discovered in influenza virus [30] and is an attractive drug target , because the process is essential for virus amplification and the enzymes involved are virus encoded . This mechanism comprises two targets: the endonuclease and the cap-binding site . However , whereas several structures of bunyavirus endonucleases have been solved and the enzyme has been characterized relatively well [7 , 16 , 18–21 , 31] , structural data on the cap-binding site of bunyaviruses is rare and functional data is missing . The publication of the structure of a reptarenavirus L protein domain structurally similar to influenza virus PB2 , but deficient of cap-binding activity raised the question whether the observed inability of cap-binding truly reflects the situation in mammarenaviruses and possibly all viruses of the Bunyavirales order [21] . Here we present the structure of a small globular domain located in the C-terminal region of RVFV L protein along with the first evidence for a functional cap-binding site within the bunyavirus L protein . The new structure is similar to influenza virus PB2 cap-binding domain and CASV putative cap-binding domain . The RVFV protein specifically binds to a cap-analogue in vitro and upon co-crystallization the cap-analogue could be detected bound between two aromatic side chains , as it is typical for several cellular and viral cap-binding proteins [25] . Still , the residues involved in the interactions with the ligand differ significantly between influenza virus PB2 and RVFV cap-binding domain . These findings pose the following questions: ( 1 ) Is a functional cap-binding domain a common feature of all viruses within the Bunyavirales order ? ( 2 ) Will it be possible to design broad-spectrum inhibitors targeting the cap-binding domain of segmented negative sense RNA viruses despite the observed differences in the binding mode of m7GTP ? The presented data support the hypothesis that bunyavirus L proteins are functionally and structurally equivalent to the concatenation of influenza virus polymerase subunits in the order PA-PB1-PB2 [17] . The presence of a β-hairpin in the RVFV structure , similar to PB2 but different from the CASV structure , suggests that the new structure might be more closely related to influenza virus cap-binding domain from an evolutionary perspective , which is contrary to what is described in the literature [32] , although it is worth mentioning that most of the phylogenetic analyses for the L gene are based on the conserved RNA-dependent RNA polymerase region . The evolutionary relation of the cap-binding domains will be difficult to prove , as the observed similarity is not obvious from the sequence level ( S6 and S7 Figs ) . Our data illustrate the structural flexibility of proteins with the same function , even though they are evolutionarily closely related . This is probably the reason why it is unlikely to reliably predict the cap-binding site in L proteins of different families within the Bunyavirales order from the sequence level . The C-terminal part of the L protein of Bunyavirales is poorly conserved , making sequence alignments challenging . Furthermore , the location of the proven or putative cap-binding domain within the polymerase proteins differs between RVFV , CASV and influenza virus , which makes a correct alignment of these sequences even more difficult . Whereas for CASV the putative cap-binding domain is between 250 and 150 amino acids distance from the C-terminus , for RVFV this domain is between 390 and 270 residues away from the C-terminus ( Fig 6A ) . The evolutionary context of this observation will be interesting to investigate in the future . We created an alignment of the proven and putative cap-binding domains of phlebo- and closely related banyangviruses , as well as arenaviruses and influenza A virus based on the observed structural similarities ( S6 Fig ) . This alignment has to be interpreted with caution , as it demonstrates the very low degree of sequence conservation in this domain ( see identity/ similarity matrix in S7 Fig ) . The lack of a reasonable alignment makes it highly unlikely to model the cap-binding domains of other bunyaviruses based on the known structures . This underlines the need to localize and solve structures of further putative cap-binding domains in order to fully understand the cap-snatching mechanism of bunyaviruses . For the development of potentially broad-spectrum inhibitors against viruses from the Bunyavirales order as well as influenza viruses , the structures of the cap-binding cavities are essential . The binding site topology of the RVFV L protein cap-binding domain , however , is different from the PB2 cap-binding site: The corresponding essential aromatic residue from the end of the long α-helix ( α1 ) in PB2 ( F404 ) is absent in the RVFV structure . Instead , this function is taken over by an aromatic residue ( F1713 ) located in the first β-strand ( β1 ) of the sheet . The equivalent residue in influenza virus PB2 ( F323 ) is responsible for stacking the ribose moiety of the cap ( S8 Fig ) . Interestingly , the aromatic residue at the end of the long α-helix ( α1 ) is present in the CASV domain ( Y1872 ) , although no reasonable candidate as a second partner for an aromatic sandwich is apparent ( Fig 6B , S8 Fig ) . We were able to demonstrate binding of the RVFV CBD to m7GTP , a cap-analogue . In a thermal stability assay we observed stabilization of RVFV CBD by m7GTP . Additionally , we determined the KD of the low affinity interaction between m7GTP and RVFV CBD , which was considerably high with 737 μM . For influenza A virus PB2 , both the shift in melting temperature in presence of m7GTP ( ΔTm <9°C , Fig 4 ) and the KD for interaction with m7GTP , which was shown to be ~1 . 5 μM in vitro [26] , point to a higher affinity interaction with the m7GTP compared to RVFV cap-binding domain . This difference is not surprising , considering the few interactions observed between the RVFV L protein fragment and the ligand in the structure ( compare S5B and S5C Fig , S2 Table ) and the fact that we are working with an isolated L protein domain . The affinity of RVFV full-length L protein for capped mRNAs has to be much higher to compete for the ligand with other cap-binding proteins inside the cytoplasm . The KD for the interaction of human eIF4E with m7GTP is ~0 . 87 μM [33] . This suggests that there have to be more residues in RVFV L protein interacting with the cap-structure or the first nucleotides of the capped mRNA apart from the small domain presented here . This is conceivable as for influenza virus PB2 residues of the cap-binding adjacent , so-called mid-link domain have been shown to interact with the capped RNA [34] . Alternatively , other mechanisms may support RVFV cap-binding activity . As an example , arenavirus Z protein has been shown to interact with eIF4E and to down-regulate its affinity for cap-structures [35] . Although most bunyaviruses do not contain a Z gene , similar scenarios involving viral proteins are conceivable . Furthermore , bunyavirus non-structural proteins have been shown to regulate host transcription and replication [36] . Additionally , cellular interaction partners of the L protein might play a role in the cap-snatching process as well as the cellular localization of viral transcription processes . For influenza virus several compounds targeting the two functions involved in cap-snatching , the endonuclease and cap-binding site , have been reported over the past years [37–41] , two of those having succeeded in clinical trials [42 , 43] . An inhibitor of influenza endonuclease has been shown to be also effective against the homologous enzyme of La Crosse bunyavirus in vitro [44] , which demonstrates the possibility to develop broad-spectrum inhibitors against the cap-cleaving endonucleases of influenza viruses and bunyaviruses . However , the observed differences in the topology of the cap-binding sites of influenza virus and RVFV are likely to be a challenge for the design of broad-spectrum inhibitors targeting cap-binding . Further structures of Bunyavirales cap-binding domains are essential to evaluate this option . The structural and functional data presented here , provide evidence for the existence of a functional cap-binding domain in bunyavirus L protein essential for viral transcription as well as the foundation for structure based drug development against RVFV cap-snatching mechanism .
Based on an alignment of phlebo- and banyangvirus L protein C-terminal sequences , we designed constructs for RVFV L protein ( strain ZH-501 , Uniprot accession: A2SZS6 ) covering the area between residues 1677 and 2008 . All sequences were cloned into pOPINF vectors [45] using the NEBuilder HiFi DNA Assembly Cloning Kit ( New England BioLabs ) . Proteins were expressed in E . coli strain BL21 Gold ( DE3 ) ( Novagen ) at 17°C overnight using TB medium and 0 . 5 mM isopropyl-β-D-thiogalactopyranosid for induction . After pelleting , the cells were resuspended in 50 mM Na-phosphate pH 7 . 5 , 100 mM NaCl , 10 mM imidazole , Complete protease inhibitor EDTA-free ( Roche ) , 0 . 4% ( v/v ) triton X-100 and 0 . 025% ( w/v ) lysozyme and subsequently disrupted by sonication . The protein was purified from the soluble fraction after centrifugation by Ni affinity chromatography . Buffers containing either 50 mM imidazole and 1 M NaCl or 50 mM imidazole and 100 mM NaCl were used for the washing steps and another buffer with 500 mM imidazole for the elution of the protein . Eluted protein was immediately diluted with 20 mM Na-phosphate pH 6 . 5 followed by removal of the N-terminal His-tag by a GST-tagged 3C protease at 4°C overnight . The protein was further purified by passing through an anion exchange chromatography column ( HiTrap Q FF , GE Healthcare ) and subsequent cation exchange chromatography ( Hi Trap SP FF , GE Healthcare , loading buffer: 50 mM Na-phosphate pH 6 . 5 , 50 mM NaCl , 10% ( w/v ) glycerol , elution with salt gradient up to 1M NaCl ) . After addition of up to 3 mM m7GTP the protein was concentrated for a final size exclusion chromatography ( Superdex 200 , 50 mM Na-phosphate , pH 6 . 5 , 150 mM NaCl , 10% ( w/v ) glycerol ) . Purified proteins were concentrated using centrifugal devices with addition of up to 6 mM m7GTP ( final concentration ) , flash frozen in liquid nitrogen , and stored in aliquots at –80°C . For thermal stability assays and isothermal titration calorimetry purification procedure was done without addition of m7GTP resulting in lower protein concentrations of max . 4 . 5 mg/ml . Protein expression was done in M9 minimal medium [46] supplemented with 1 mM MgSO4 , 0 . 4% glucose , 0 . 0005% thiamine and 200 μM FeSO4 at 17°C overnight . Incorporation of seleno-methionine was achieved by metabolic inhibition of methionine biosynthesis in E . coli prior to addition of seleno-methionine and induction with 1 mM isopropyl-β-D-thiogalactopyranosid . Cells were harvested and the labelled protein was purified as described but in presence of 5 mM β-mercaptoethanol for Ni affinity purification and 10 mM DTT for the remaining purification steps . RVFV CBD13 protein was produced with seleno-methionine labelling , CBD9 only as native protein . Protein crystals of CBD9 grew at 8 mg/ml protein concentration in 21% PEG 4000 , 10% glycerol , 12% isopropanol and 100 mM Na-Citrate pH 6 . 0 in a sitting drop vapor diffusion setup at 20°C . Protein crystals of CBD13 grew at 9 mg/ml protein concentration in 24% PEG 2000 MME , 200 mM Trimethylamine N-oxide , 2 mM TCEP , 5 mM dithiothreitol , 2 mM m7GTP and 100 mM Tris , pH 8 . 5 in a sitting drop vapor diffusion setup at 20°C . Crystals were flash frozen in liquid nitrogen without cryo protectants in case of CBD13 and with 25% ethylenglycol in reservoir solution for CBD9 . Datasets were obtained at beamline P13 of PETRA III at Deutsches Elektronen Synchrotron ( DESY ) , Hamburg , Germany . Datasets were processed with iMosflm [47] and the RVFV CBD13 structure was solved by the single anomalous dispersion method using PHENIX AutoSol [48] . The structure was refined by iterative cycles of manual model building in Coot [49] and computational optimization with PHENIX [48] . Visualization of structural data was done using the PyMOL Molecular Graphics System , Version 1 . 7 Schrödinger , LLC . Thermal stability of RVFV CBD13 was measured by thermofluor assay [50] . The assay contained a final concentration of 15 μM of CBD13 protein , 20 mM Na-phosphate pH 6 . 5 , 100 mM NaCl , SYPRO-Orange ( final dilution 1:1000 ) and either no additive or between 1 and 10 mM of m7GTP , m7GpppG , GTP or ATP . Thermal stability of influenza A virus PB2 cap-binding domain was assessed at a final protein concentration of ~10 μM in the same setup . Affinity of RVFV CBD13 to m7GTP and GTP was measured by isothermal titration calorimetry ( ITC ) using a MicroCal PEAQ-ITC instrument ( Malvern Panalytical ) . The instrument was calibrated using a control reaction of Ca2+ binding to EDTA as supplied by Malvern Pananalytical . Proteins were dialyzed overnight at 4°C against 50 mM Na-phosphate , pH 6 . 5 , 150 mM NaCl , 10% ( w/v ) glycerol . Ligand m7GTP was dissolved and GTP was diluted in the exact same dialysis buffer . Titrations were done with slightly different setups: ( 1 ) 128 μM CBD13 in the cell and 3 . 8 mM m7GTP in the syringe at 20°C with 13 injections of 3 μl ( first injection 0 . 5 μl ) , ( 2 ) 194 μM CBD13 in the cell and 9 . 8 mM m7GTP in the syringe at 25°C with 19 injections of 2 μl ( first injection 0 . 5 μl ) , or ( 3 ) 194 μM CBD13 in the cell and 9 . 8 mM GTP in the syringe at 25°C with 19 injections of 2 μl ( first injection 0 . 5 μl ) . Spacing between injections was constant with 150 seconds for all measurements . Data were analyzed and fitted with the respective PEAQ ITC evaluation software ( Malvern ) applying a single site binding model and fixing the stoichiometry value to 1 . The experiments were performed in the context of the T7 RNA polymerase-based RVFV ambisense minireplicon system essentially as described by Jérôme et al . [23] . L gene mutants were generated by mutagenic PCR using pCITE-L as a template . The PCR products containing the functional cassette for expression of mutant L protein were purified , quantified spectrophotometrically , and used for transfection without prior cloning . The presence of the artificial mutation was ascertained by sequencing . BSR-T7/5 cells stably expressing T7 RNA polymerase [51] ( kindly provided by Ursula Buchholz and Karl-Klaus Conzelmann ) were transfected per well of a 24-well plate with 250 ng of L gene PCR product , 750 ng of RVFV minigenome plasmid expressing Renilla luciferase ( Ren-Luc ) , 500 ng of pCITE-NP expressing NP , and 10 ng of pCITE-FF-Luc expressing firefly luciferase as an internal transfection control . One day after transfection , total RNA was purified using an RNeasy Mini Kit ( Qiagen ) for northern blotting or cells were lysed in 100 μl of passive lysis buffer ( Promega ) per well , and assayed for firefly luciferase and Ren-Luc activity using the dual-luciferase reporter assay system ( Promega ) . Ren-Luc levels were corrected with the firefly luciferase levels ( resulting in standardized relative light units [sRLU] ) to compensate for differences in transfection efficiency or cell density . For northern blot analysis , 500 ng of RNA was separated in a 1 . 5%-agarose–formaldehyde gel and transferred onto a Roti-Nylon plus membrane ( Roth ) . Blots were hybridized with a 32P-labelled antisense riboprobe targeting the Ren-Luc gene , and RNA bands were visualized by autoradiography using a Typhoon scanner ( GE Healthcare ) . To verify expression of L protein mutants , BSR-T7/5 cells in a well of a 24-well were transfected with 500 ng of PCR product expressing L protein mutants tagged at the C-terminus with a 3xFLAG sequence . To enhance the expression level , the cells were additionally inoculated with Modified Vaccinia virus Ankara expressing T7 RNA polymerase ( MVA-T7 ) [52] . Cytoplasmic lysate was separated in a 3–8% Tris-acetate polyacrylamide gel , transferred to nitrocellulose membrane ( Whatman ) , and detected by immunoblotting using peroxidase-conjugated anti-FLAG M2 antibody ( 1:10 , 000 ) ( A8592; Sigma-Aldrich ) . L protein bands were visualized by chemiluminescence using SuperSignal West Femto substrate ( Pierce ) and a FUSION SL image acquisition system ( Vilber Lourmat ) . | Rift Valley fever virus ( RVFV ) is endemic to sub-Saharan Africa and the Arabian Peninsula and leads to abortions in and death of ruminants . The virus can also be transmitted to humans causing febrile illness up to hemorrhagic fever with the possibility of fatal outcome . As there is currently no human vaccine or specific treatment available and because of the high epidemic potential , WHO has listed RVFV on its R&D Blueprint for urgent development of medical countermeasures . In order to amplify , the virus needs to transcribe and replicate the viral genome inside the cell cytoplasm . For transcription , the virus uses a process called cap-snatching , which is essentially depending on two functions presumed to reside within the large viral L protein: the ability to bind cap-structures and the activity of cleaving them off from cellular mRNA . Both functions could serve as specific targets for antiviral drug design . We identified and solved the structure of the cap-binding domain of RVFV and provide the first evidence for the presence of a functional cap-binding site in the L protein of bunyaviruses . Comparison with cap-binding proteins of related viruses revealed similarities and important differences critical for the development of potential broad-spectrum antivirals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"rift",
"valley",
"fever",
"virus",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"viral",
"structure",
"orthomyxoviruses",
"viruses",
"rna",
"viruses",
"protein",
"structure",
"bunyaviruses",
"resear... | 2019 | Structure of a functional cap-binding domain in Rift Valley fever virus L protein |
Segmentation of the vertebrate body axis is initiated through somitogenesis , whereby epithelial somites bud off in pairs periodically from the rostral end of the unsegmented presomitic mesoderm ( PSM ) . The periodicity of somitogenesis is governed by a molecular oscillator that drives periodic waves of clock gene expression caudo-rostrally through the PSM with a periodicity that matches somite formation . To date the clock genes comprise components of the Notch , Wnt , and FGF pathways . The literature contains controversial reports as to the absolute role ( s ) of Notch signalling during the process of somite formation . Recent data in the zebrafish have suggested that the only role of Notch signalling is to synchronise clock gene oscillations across the PSM and that somite formation can continue in the absence of Notch activity . However , it is not clear in the mouse if an FGF/Wnt-based oscillator is sufficient to generate segmented structures , such as the somites , in the absence of all Notch activity . We have investigated the requirement for Notch signalling in the mouse somitogenesis clock by analysing embryos carrying a mutation in different components of the Notch pathway , such as Lunatic fringe ( Lfng ) , Hes7 , Rbpj , and presenilin1/presenilin2 ( Psen1/Psen2 ) , and by pharmacological blocking of the Notch pathway . In contrast to the fish studies , we show that mouse embryos lacking all Notch activity do not show oscillatory activity , as evidenced by the absence of waves of clock gene expression across the PSM , and they do not develop somites . We propose that , at least in the mouse embryo , Notch activity is absolutely essential for the formation of a segmented body axis .
Segmentation is a key feature of the body plan of all vertebrates , including humans , that initiates very early in embryonic development . The first sign of metamerism or segmentation is seen when vertebrate embryos develop somites , the precursors of several segmented organs such as the axial skeleton , body skeletal muscles and part of the dermis . Somites are formed in a highly regulated process called somitogenesis from the unsegmented presomitic mesoderm ( PSM ) [1]–[3] . During the formation of somites the most mature PSM cells located at the rostral end of the PSM bud off as an epithelial sphere of cells to form the somite . Somite formation occurs simultaneously with the recruitment of newly generated mesenchymal cells from the primitive streak/tail bud into the caudal region of the PSM [2]–[4] . Critical molecular and embryological experimental data obtained in the last ten years has shown that somitogenesis is governed by a molecular oscillator [5] that drives cyclic expression of genes in the PSM and which is coupled to the formation of the somites [2] , [3] , [6]–[8] . Expression of these cyclic genes is coordinated such that a wave of expression travels caudo-rostrally throughout the PSM during the formation of one somite . All cyclic genes identified to date encode either ( a ) components or modulators of the Notch pathway ( b ) components of the Wnt pathway or ( c ) components of the FGF pathway [2] , [3] , [6]–[8] . There are discrepancies in the literature regarding the role ( s ) of Notch signalling during the process of somite formation . At least in the zebrafish embryo it seems clear that Notch signalling has a predominant function in the synchronization of clock gene oscillations , where inhibition of Notch is not sufficient to interrupt the generation of a segmented body plan [7]–[11] . This view of Notch as a clock synchronizer has also been proposed to operate during mouse somitogenesis [12] , [13] . On the other hand , data generated in chick , mouse and zebrafish is consistent with Notch being an important component of the molecular oscillator in different vertebrate species . Thus , ectopic expression of Lfng in the chick PSM or morpholino treatment against her genes in zebrafish embryos interferes with cyclic gene expression and leads to the generation of irregular somites , similar to the phenotype observed in different mouse and zebrafish transgenic lines carrying a mutation in various components of the Notch pathway [2] , [3] , [8] . Finally , a third possibility is that Notch signalling may have dual functions as both a clock generator as well as a clock synchronizer [14] , [15] . In this report we re-examine the implication of Notch signalling in the mechanism of the mouse somitogenesis oscillator and in murine somite formation by analysing embryos carrying a mutation in different components of the Notch pathway , such as Lunatic fringe ( Lfng ) , Hes7 , Rbpj and presenilin1/presenilin2 ( Psen1/Psen2 ) , and by pharmacological blocking of the Notch pathway . Our results show that , at least in the mouse embryo , Notch activity , be it cyclic or non-cyclic , is critically required both for the generation of periodic transcription of cyclic genes by the somitogenesis oscillator and for the formation of the somites .
To further clarify the role of Notch signalling during mouse somitogenesis we decided to analyse in detail the phenotype of two mouse knockout lines , namely Hes7−/− [16] and Lfng−/− [17] , [18] , two components of the Notch pathway . Hes7 is a downstream target of Notch and encodes a repressor of transcription previously shown to be a negative component of the machinery of the somitogenesis oscillator [19] , [20] . Lfng is also a downstream target of Notch that encodes a glycosyltransferase that modulates the potential of the Notch receptor protein to interact with its ligands Delta and Serrate/Jagged [21] . We first analysed the organization of the somites in E9 . 5–10 . 5 homozygous null embryos of both mutant lines , Lfng−/− ( n = 5 ) and Hes7−/− ( n = 6 ) , and observed that they are irregular , their size is not uniform , they are occasionally fused ( Figure 1A–1C ) and sometimes they display left-right asymmetry ( data not shown ) . We observed these disordered somites in the mutant embryos over a variety of developmental stages , E8 . 5–10 . 5 ( data not shown ) , as previously reported [16] , [20] . To visualise the long term impact of the absence of these two Notch-related components of the oscillator in the formation of the metameric body plan we performed alcian blue/alizarin red staining , which stains all bones and cartilage , using E18 . 5 mouse embryos . As expected , Lfng−/− ( n = 4 ) and Hes7−/− ( n = 4 ) embryos displayed skeletal abnormalities along the spinal column ( Figure 1D–1F; [16]–[18] ) . Thus , in the absence of Lfng or Hes7 the process of somitogenesis is not properly regulated . However , of more relevance for this particular analysis is the fact that these embryos are capable of producing somites and vertebrae at all , which indicates the potential existence of periodic activity produced by the somitogenesis oscillator machinery . To address the nature of this oscillator we analysed the expression of a number of oscillatory genes . Initially we analysed the expression of Notch-regulated cyclic genes . The expression of Hes7 was upregulated in the entire PSM of Lfng−/− embryos ( n = 10 , Figure 1I ) . Similarly , we analysed Hes7−/− embryos and observed that they also displayed upregulated expression of Lfng ( n = 20 , Figure 1L; [16] ) . Thus , the results show that in the absence of important negative regulatory components , such as Hes7 and Lfng , Notch activity appears upregulated even if the mutant embryos are still able to generate segmented structures . In principle , the observed upregulation of Notch downstream targets in the Lfng−/−embryos could culminate in an accumulation of the mRNA for these genes along the PSM . To test this , we decided to measure the amount of Hes7 mRNA present in the rostral half of the PSM of wild type or Lfng−/− embryos . To that end we isolated the total RNA from pooled rostral half PSM samples of several wild type and mutant embryos of unknown cyclic phases and then performed quantitative RT-PCR . We observed that the relative expression level of Hes7 revealed no statistically significant difference between wild type ( n = 12 ) and Lfng−/− ( n = 10 ) PSM samples ( Figure 2A; t-test , df = 20 , P = 0 . 130 ) . One possible explanation for this lack of accumulation of Hes7 mRNA in the Lfng−/− embryos might be that the transcripts are in fact produced and degraded as in the wild type embryo , albeit not quite as efficiently . We decided to re-examine Hes7 mRNA expression in these Lfng−/− embryos in more detail . When we re-analysed Lfng−/− embryos with the Hes7 probe carefully monitoring the intensity of the revelation step we observed different patterns or phases of expression ( n = 13 , Figure 2B and 2C; [22] ) . Longer staining of the same mutant embryos led to the general upregulation of Hes7 described above ( Figure 2B' and 2C' , Figure 1I ) . Under similar conditions of longer staining this general upregulation is not observed using wild type embryos ( n = 25 , data not shown ) . To further corroborate these data we analysed Lfng−/− embryos using a Hes7 intronic probe in order to detect nascent pre-spliced mRNA and thereby to show the location of active transcription [23] . The Lfng−/− embryos ( n = 6 ) presented patterns of Hes7 expression similar to those observed in wild type embryos ( n = 6 , Figure 2D and 2E , data not shown ) . To confirm that these different patterns corresponded indeed to a dynamic activity we performed a half embryo analysis , in which the tail of an embryo is split longitudinally in two halves , then one half is immediately fixed and the other is cultured for 60 minutes before fixation [24] , [25] . In situ hybridisation with an intronic Hes7 probe on samples prepared using this type of analysis showed that the two halves displayed different patterns of expression ( n = 5 , Figure 2F ) , which clearly indicates that in the absence of Lfng activity the expression of the Notch-related cyclic gene Hes7 is still dynamic . Similarly , when we analysed the expression of a second Notch-related cyclic gene , Nrarp , we also found different patterns of expression similar to those observed in wild type embryos ( n = 8 and n = 12 respectively , Figure 2G and 2H , Figure 3C and 3D; [26] ) . In addition , we analysed the expression of Hes7 protein in tails of E10 . 5 Lfng−/− embryos ( n = 8 ) using a specific anti-Hes7 antibody [19] and found that Hes7 protein also displayed different phases of expression , consistent with it being cyclic , although the boundaries of expression were not as sharp as in the wild type ( n = 8 and n = 10 respectively , Figure 2I and 2J; data not shown ) . Finally , we tested the expression of the cleaved intracellular portion of Notch ( NICD ) , which is the active fragment of Notch and is generated when Notch receptor , after its interaction with the ligand , is processed by the γ-secretase complex . Once NICD is produced it translocates to the nucleus where it binds to RBPj and the complex becomes a transcriptional activator of downstream targets [21] . NICD has previously been reported to display a dynamic expression profile in the PSM of wild type mice embryos [27] , [28] . Using an antibody specific for NICD we stained tail sections of E10 . 5 Lfng−/− embryos and observed different patterns of NICD expression ( n = 9 , Figure 2K and 2L ) , consistent with Notch activity still being dynamic . These results clearly indicate that in the absence of the glycosyltransferase Lfng it is possible to detect dynamic NICD and dynamic expression of Notch-related cyclic genes , although not with sharp boundaries of expression , which suggests that Lfng could be an important but not a critical component of the mouse oscillator . In order to examine more closely whether the Hes7−/− embryos also retained some cyclic Notch activity , we first analysed these embryos by in situ hybridisation with an intronic probe against the Notch-regulated cyclic gene Lfng . Strikingly , we found a different situation in the Hes7−/− embryos than we had observed in the Lfng−/− embryos and we were not able to detect the existence of different phases of Lfng expression . During the course of a controlled staining all Hes7−/− embryos ( n = 10 ) displayed the same profile of a broad rostro-caudal gradient of expression ( Figure 3B ) . Longer staining of the same mutant embryos led to the general upregulation of Lfng described above ( Figure 3B' , Figure 1I ) . Similarly , the expression pattern of the Notch-regulated cyclic gene Nrarp was identical in all Hes7−/− embryos ( n = 7 , Figure 3E ) . These non-dynamic patterns are consistent with the fact that in these Hes7−/− embryos we did not see different patterns of NICD expression either , rather it was detected in a rostro-caudal gradient of expression across the PSM ( n = 6 , Figure 3F ) . Expression of the non-cyclic Notch target gene Mesp2 [29] , [30] was retained in the rostral region of the PSM of Hes7−/− embryos , although the expression domain was not as sharp as in wild type embryos ( Figure 3G and 3H; [16] ) . It has been reported that Mesp2 expression is severely compromised in Dll1−/− and Rbpj−/− mutant embryos [31] . Thus , the Mesp2 band of expression observed in the Hes7−/− embryos is probably due to the presence of non-dynamic Notch signalling activity in the PSM . In summary , there is a non-dynamic expression of Notch-based cyclic genes in the PSM of Hes7−/− embryos , which mirrors the non-dynamic Notch activity in this tissue . Since in the Hes7−/− embryos , unlike the Lfng−/− embryos , there does not appear to be cyclic activity of Notch-regulated cyclic genes we investigated whether any of the FGF or Wnt regulated cyclic genes were oscillating in the PSM of these Hes7−/− embryos . Based on the patterns of expression observed in Dll1−/− embryos it has been proposed that the expression of Axin2 is independent of Notch activity [32] . We first examined the expression of the Wnt-related cyclic gene Axin2 in the PSM of Hes7−/− embryos ( n = 30 ) and observed different patterns of expression ( Figure 4D and 4E; [20] ) . In addition , a fix and culture analysis clearly indicated that this Axin2 expression is dynamic in the Hes7−/− mutant background ( n = 21 , Figure 4F ) . We also analysed the expression of the FGF/Wnt-regulated gene Snail1 by in situ hybridisation ( n = 10 ) and fix and culture ( n = 5 ) and observed that it is also dynamic ( Figure 4J–4L; [33] ) . Our results indicate that the Hes7−/− embryo makes irregular somites in the absence of cyclic Notch but in the presence of cyclic Wnt activity . The third pathway described to be a critical component of the murine somitogenesis oscillator is the FGF pathway . FGF is reported to be responsible for the initiation of Hes7 expression in the caudal PSM [34]–[36] and some components of the FGF pathway , such as Dusp4/6 and Sprouty2/4 , have been shown to display dynamic expression in the PSM [34] , [37] , [38] . We tested the expression of Dusp6 ( n = 30 ) and Sprouty2 ( n = 32 ) in Hes7−/− embryos and observed different patterns of expression , consistent with dynamic FGF activity in the PSM of these embryos ( Figure 4O , 4P , 4S , and 4T ) . In summary , we conclude that embryos lacking Hes7 retain dynamic activity of the Wnt-regulated genes and FGF-regulated genes of the somitogenesis oscillator , which is likely to underlie the generation of periodicity and the formation of irregular somites in these embryos . Our data indicate that in Lfng−/− embryos the cyclic gene oscillations comprise elements of the Notch pathway ( Figure 2 ) . This is in contrast to the situation in the Hes7−/− embryos because the oscillatory mechanism appears to be based only on Wnt and FGF-dependent genes . It is not clear , however , if this Wnt/FGF-based oscillator is sufficient to generate segmented structures , such as the somites , in the complete absence of all Notch activity , since in these Hes7−/− embryos there remains clear evidence of non-dynamic Notch activity , which could be a critical requirement for the proper function of an Wnt/FGF-based somitogenesis oscillator and/or the formation of the somites . To more definitively test the relevance of Notch activity during the process of somitogenesis we decided to re-analyse homozygous null embryos from two other mutant lines widely accepted to develop in the complete absence of Notch activity , Rbpj−/− and Psen1−/−;Psen2−/− [39]–[41] . We first evaluated the situation in Rbpj−/− embryos . RBPj is the transcriptional repressor to which NICD binds in the nucleus in order to activate expression of downstream target genes [21] , [42] , [43] . Rbpj−/− embryos die at approximately E9 . 5 after forming a variable number ( zero to five ) of disorganized and irregular somite-like structures [39] . As expected , we observed that at stage E8 . 5–9 . 0 the expression of the Notch-regulated cyclic gene Lfng ( n = 6 ) was lost in the PSM ( Figure 5B ) . Barrantes and colleagues have previously reported that in a few cases they were able to detect a single faint stripe of Lfng in the rostral PSM [31] . Surprisingly , however , we found that the Notch-related cyclic gene Hes7 was still present along the PSM of these Rbpj−/− embryos ( n = 5 , Figure 5D and 5E; [34] ) . In fact , Hes7 expression can be detected with different patterns of expression in a broad caudal domain and in restricted bands in the rostral region , similar to the expression observed in wild type embryos ( Figure 5C ) , suggesting that its dynamic character may still be functional in these mutant embryos . Similarly we also found that the Wnt/FGF-based cyclic gene Snail1 displayed different patterns of expression , including both a caudal domain and a rostral band of expression ( n = 14 , Figure 5I and 5J; [33] ) . The Wnt-related cyclic gene Axin2 ( n = 4 ) and the FGF-related cyclic genes Dusp6 ( n = 4 ) and Sprouty2 ( n = 5 ) were also found expressed along the PSM with patterns of expression similar to those found in wild type embryos ( Figure 5G , 5L , and 5N ) . Of critical importance to this study is the fact that in Rbpj−/− embryos two Notch-dependent cyclic genes , Lfng and Hes7 , respond differently to the absence of RBPj activity and at least Hes7 displays patterns of expression in the PSM similar to those observed in the wild type , which may be dynamic . These data raise the question of whether these mutant embryos do in fact develop in the complete absence of Notch activity or whether there remains some residual RBPj-independent Notch activity similar to what has been described in Drosophila [44] , [45] . Feller and colleagues have shown that the expression of these two Notch targets , Lfng and Hes7 , is also differentially affected following perturbation of Notch activity in the PSM of embryos expressing constitutive-activate Notch ( T-NICD ) [13] . Thus , it would appear that they are not equally sensitive to the levels of NICD as an input to their expression . It is formally possible that the absence of RBPj results in a severe decrease of Notch signalling leading to a loss of Lfng but that a certain level of RBPj-independent NICD activity remains , which could act to maintain the expression of Hes7 . To further investigate this possibility we decided to explore if NICD is expressed in the PSM of the Rbpj−/− embryos . Our immunostaining on sections clearly indicate that indeed it is possible to detect weak NICD expression in the PSM of these mutant embryos ( Figure 5P; [46] ) . From these two results , the existence of different patterns of expression of Hes7 in the PSM and the expression of NICD , we conclude that the Rbpj−/− mutant line is not appropriate to definitively test the significance of a complete lack of Notch activity during the process of somitogenesis . We next examined embryos from the double knockout line Psen1/Psen2 , which generate mutant embryos lacking all presenilin activity [40] , [41] . Presenilin is the catalytic component of the γ-secretase complex responsible for the cleavage of Notch receptor and the generation of NICD . In principle , this mutant mouse line should lack all Notch activity . Psen1+/−;Psen2−/− embryos retaining one allele of presenilin1 did not display a somitic phenotype ( n = 20 , Figure 6A ) and showed normal patterns of Hes7 expression along the PSM ( Figure 6C and 6D ) . In contrast , Psen1−/−;Psen2−/− embryos lacking the two presenilins failed to form any somites ( n = 22 , Figure 6B; [40] ) and Hes7 expression was absent in the medial and rostral PSM , consistent with this expression domain being entirely Notch dependent . Nevertheless , it was possible to detect weak Hes7 expression in the tail bud region ( n = 4 , Figure 6E ) , which may be indicative of FGF-induced activation in this domain since FGF is reported to be responsible for the initiation of Hes7 expression in the caudal PSM [34] , [35] . Consistent with the idea that these double mutant embryos develop in the complete absence of Notch activity is the fact that NICD was not detected by immunostaining on sections prepared from these double mutant embryos ( Figure 6J ) . To confirm this negative result we also performed western blot analysis using protein samples prepared with embryonic fibroblasts from the Psen1−/−;Psen2−/− embryos [47] and observed that NICD is not produced ( Figure 6K; [48] ) . These data indicate that , in contrast to our observations in the Rbpj−/− embryos , the Psen1−/−;Psen2−/− embryos develop in the complete absence of Notch activity , as judged by the absence of NICD production and the absence of cyclic expression of Notch-dependent Hes7 . Interestingly , we also found that in these double mutant embryos the expression of Axin2 ( n = 4 ) , Snail1 ( n = 4 ) , Dusp6 ( n = 4 ) and Sprouty2 ( n = 4 ) is lost along the PSM and when they are detected the remaining staining is restricted to the neural tube or caudal tail bud ( Figure 6F–6I ) . Together these results indicate that the Notch activity appears to be critical both to maintain any form of dynamic clock gene expression in the PSM and to generate somites . We decided to confirm these observations by blocking all Notch activity in the PSM using a pharmacological treatment and then analysed the effect of this treatment on the expression of oscillatory genes and somite formation , similar to the approach used in recent studies focussing on the role of Notch during somitogenesis in the zebrafish embryo [6] , [10] , [11] , [49] . We first performed the treatment using the half embryo assay in the presence or absence of either 100 µM DAPT or 100 nM LY411575 , two reagents that block Notch by inhibiting the γ-secretase cleavage of the Notch receptor [50]–[54] . We confirmed by western blot that after a 3 hour exposure to the Notch-blocking drugs the expression of NICD was abrogated ( Figure 6K ) , as previously reported [48] , [55] . Furthermore the expression of Lfng and Nrarp was completely abolished from the PSM , as expected ( n = 8 , Figure 7A; [24] , [25]; Wright et al . , submitted ) . In addition , the drug-treated samples showed no expression of Hes7 in the rostral PSM , consistent with this expression domain being entirely Notch dependent . In the rest of the PSM these treated explants also showed either no expression of Hes7 at all or only a restricted weak caudal domain in the tail bud ( n = 14 , Figure 7D ) , which is reminiscent of the Hes7 pattern of expression observed in the Psen1−/−;Psen2−/− embryos ( Figure 6E ) . As mentioned above , Hes7 expression in the caudal PSM is initiated by FGF activity [34] , [35] . Therefore , if Notch and FGF activities are blocked simultaneously Hes7 expression should disappear completely from the PSM . Consistent with this idea , when half embryo samples were cultured in the presence of both DAPT and SU5402 , a drug that blocks FGF signalling , the expression of Hes7 was completely abolished in the caudal PSM ( n = 10 , Figure 8G ) . Drug treatment with DAPT/LY411575 also drastically reduced Axin2 expression throughout the PSM as compared to untreated controls , and when detected it was restricted to a weak caudal domain ( n = 14 , Figure 7G ) . These data indicate that drug treatment blocked the rostral progression of cyclic expression of both the Notch-based gene Hes7 , as well as that of the Wnt-regulated gene Axin2 across the PSM . To test the effect of DAPT/LY411575 treatment on somite formation we cultured E8 . 0–8 . 5 wild type mouse embryos for 18–20 hours using an in vitro roller culture system in the presence or absence of the drug . During this period of time control embryos formed 9–10 extra somites ( n = 46 , Figure 7B , 7E , 7H , 7K , 7N , and 7Q ) , however embryos treated with Notch-blocking drugs formed a very limited number of somites , varying from 0 to 2 ( n = 65 , Figure 7C , 7F , 7I , 7L , 7O , and 7R ) . As expected , drug-treated embryos did not show Lfng expression ( Figure 6C ) . In addition , none of the treated embryos expressed Hes7 in the rostral or medial PSM and in those that did show expression it was very weak and restricted to the caudal end of the tail bud ( n = 9 , Figure 7F ) . Likewise , none of these treated embryos expressed Axin2 in the rostral PSM but a proportion showed some weak expression in the caudal domain ( n = 8 , Figure 7I ) . To test if FGF-dependent cyclic genes are also affected under these conditions we evaluated the expression of Snail1 , Dusp6 and Sprouty2 and found that their expression seems to be not affected following a 3 hour exposure to Notch-blocking drugs ( n = 10 , n = 8 and n = 9 respectively , Figure 7J , 7M , and 7P ) , but they were severely downregulated in the PSM after overnight incubation with the drugs ( n = 8 , n = 8 and n = 11 respectively , Figure 7L , 7O , and 7R ) . Thus , these results are consistent with those obtained by analysis of the Psen1/Psen2 double mutant embryos and show that , at least in the mouse , Notch activity is critical for both the maintenance and rostral progression of oscillations of the segmentation clock along the PSM and for somite formation . Because of the oscillatory nature of cyclic gene expression in the caudal PSM of normal non-treated embryos it is possible that the remaining expression of these genes observed in the tail bud region after treatment with the drugs is indicative of the existence of an FGF/Wnt based pacemaker operational and dynamic in the absence of Notch activity . Under this paradigm the oscillations might be generated in the primitive streak/tail bud by a Notch-independent mechanism and Notch would then act to propagate these oscillations along the PSM . To test the possible existence of this Notch-independent pacemaker we used the half embryo assay and treated both halves with DAPT or LY411575 for 3 hours , then fixed one half and cultured the second half for 1 additional hour before fixation . When we performed this analysis we found that the expression of Lfng was completely absent in both halves , as expected ( n = 3 , Figure 8A ) . Under these conditions , if the expression of cyclic genes , such as Hes7 , continues to oscillate in the caudal PSM after drug treatment we should find pairs of samples in which the expression domain is different in the two sides such that the cyclic gene is present in one half and absent in the other . When we analysed Hes7 we found that the expression was weak in the two halves as compared with the intensity observed in control non-treated samples , as described above . By extending the colour revelation step we observed that Hes7 expression was present and restricted to the caudal region of the PSM of drug-treated samples ( n = 8 , Figure 8B ) . Similarly , Axin2 displayed a weak expression profile restricted to the caudal end of the PSM ( n = 9 , Figure 8C ) . In addition , we observed that Snail1 , Dusp6 and Sprouty2 displayed similar patterns of expression in both halves ( n = 12 , n = 11 and n = 14 respectively , Figure 8D–8F ) , indicating that their expression was also not dynamic . In summary , all these results taken together demonstrate that , in the mouse embryo , in the absence of Notch signalling the expression of cyclic genes is lost in the rostral and medial PSM and is dramatically reduced in the caudal PSM where it loses its dynamism , and this is not consistent or compatible with the existence of a Notch-independent pacemaker .
In this study we have investigated the implication of Notch signalling in the mechanism of the mouse somitogenesis oscillator . We found that in the absence of Lfng the mouse embryo is still able to display dynamic expression of all cyclic genes analysed . In the absence of Hes7 , however , only the expression of FGF and Wnt-regulated cyclic genes is still dynamic in a PSM that displays non-cyclic Notch activity along its extension . Surprisingly , in the absence of RBPj there is still some RBPj-independent Notch activity , as evidenced by the expression profile of Hes7 . Our data show that the double mutant embryos Psen1−/−;Psen2−/− develop in the complete absence of Notch activity and they do not form somites or display oscillatory gene expression , as evidenced by the lack of expression of cyclic genes along the PSM . Similar defects are produced in wild type embryos cultured in the presence of Notch-blocking drugs . We propose that , contrary to what happens during zebrafish development , in the mouse embryo Notch activity , cyclic or non-cyclic , is critically required both for the maintenance of the somitogenesis oscillator and for the formation of the somites ( Figure 8H ) . It has been previously shown that Lfng and Hes7 are two important components of the Notch pathway and interfering with their functions seems to affect the somitogenesis oscillator [19] , [20] , [24] , an idea supported by the phenotype displayed by Lfng−/− and Hes7−/− embryos [17]–[19] . Homozygous mutant embryos have clear somitic abnormalities that later in development result in skeletal malformations of vertebrae and ribs . However , the fact that these mutant embryos make somites at all indicates that the somitogenesis oscillator may still be producing oscillations and generating periodicity in the absence of these two proteins . A first analysis suggested that the Notch pathway was upregulated along the entire PSM of Lfng−/− embryos , as judged by the expression of Notch-dependent cyclic genes , although a more careful analysis revealed dynamic Hes7 expression . The existence of this dynamism in the PSM of the Lfng−/− embryos is corroborated by the results obtained by immunostaining with the anti-Hes7 and anti-NICD antibodies that demonstrated there is periodic production of Hes7 mRNAs and periodic production of NICD . Based on the patterns of expression described in the literature the induction of all Notch target genes in the PSM is largely generated by the interaction of Notch and the Delta family of ligands [18] , [31] , [56]–[59] . Our data indicates that in the mouse PSM Delta-driven Notch signalling can occur in the complete absence of Fringe activity with an intensity equivalent to when Fringe is present . We cannot exclude , however , the formal possibility that in the PSM of the Lfng−/− embryos another enzyme can substitute for Lfng in order to maintain the activity of the Notch pathway . Further analysis will be required to clarify the situation . In contrast to Lfng−/− embryos , Hes7−/− embryos do not show cyclic expression of the Notch downstream target genes Lfng and Nrarp , and NICD is expressed in a rostro-caudal gradient , which is not consistent with Notch signalling being dynamic . On the other hand , the fact that the expression of FGF/Wnt-based cyclic genes is still dynamic in the Hes7−/− PSM indicates the FGF/Wnt-components of the somitogenesis oscillator are still operating in these embryos , which is likely to underlie the generation of periodicity and the formation of irregular somites in these embryos . A similar explanation may underlie the existence of somites in a transgenic mouse line that expresses activated non-cyclic Notch throughout the PSM ( T-NICD ) and shows no dynamic expression of Notch genes but , nevertheless , the cyclic expression of Axin2 is unaffected [13] . Thus , in both Hes7−/− embryos and in the T-NICD transgenic embryos [13] Wnt-based genes continue to oscillate in a background of non-cyclic Notch activity . It remains unclear whether the presence of somites and the existence of cyclic gene expression in these two genetic backgrounds , Hes7−/− and T-NICD , are due to the combination of non-cyclic Notch together with cyclic Wnt signalling or whether cyclic Wnt alone is sufficient to account for this . The analysis of Rbpj−/− embryos , thought to develop in the absence of Notch activity , failed to resolve this issue since , surprisingly , the results show that at least one Notch-related cyclic gene , Hes7 , is still expressed with different patterns of expression along the PSM suggesting its expression is still dynamic . While it remains a formal possibility that this residual Hes7 expression is Notch-independent the fact that we report NICD is present at low levels in the PSM of the Rbpj−/− embryos strongly support that the Hes7 expression is a consequence of a poorly defined RBPj-independent Notch activity in this tissue [21] . When we analysed the phenotype of the Psen1−/−;Psen2−/− embryos we found that they do not display any kind of activity of the segmentation clock , as suggested by the lack of different patterns of expression of different Notch-based , Wnt-based and FGF-based cyclic genes , and because they do not form somites . We also found that NICD is not produced in these embryos , a clear indication that these double mutant embryos really develop in the absence of Notch signalling . We can not rule out the formal possibility that there is a Notch-independent γ-secretase activity implicated in segmentation [27] . However , we think this is unlikely because amongst the list of type I transmembrane proteins known to be substrates for γ-secretase [60] , [61] only the Notch components have been described to be implicated in the oscillator involved in somitogenesis . In addition , when we treated samples from wild type mouse embryos in vitro with drugs that block Notch cleavage we also inhibited the dynamic expression of all cyclic genes and the generation of somites beyond those already determined or in the process of being formed in the rostral PSM at the time of treatment ( 1 or 2 ) . Thus , the generation of temporal and spatial periodicity in the mouse embryo absolutely requires Notch activity . In the absence of all Notch activity no oscillations occur and no somites are formed . Feller et al . recently reported that NICD is not detected in the PSM of Pofut1−/− embryos , a mutant mouse line carrying a mutation in another relevant component of the Notch pathway , although these embryos are nevertheless still able to generate a significant number of irregular somites [13] . The authors concluded that these Pofut1−/− mutant embryos develop and generate somites in the absence of Notch signalling . One explanation for the difference in their interpretation compared to ours is the detection limit of our respective assays for NICD . Indeed the expression of NICD we detect in Rbpj−/− embryos , which also develop a limited number of somites , is very weak and could easily be overlooked . As yet an analysis of the expression profile of different Notch-based cyclic genes , including Lfng and/or Hes7 , has not been performed with the Pofut1−/− embryos . Further investigation will be required to clarify these discrepancies . The only Hes7 expression domain remaining in the Psen1−/−; Psen2−/− embryos or in wild type embryos after treatment with Notch-blocking drugs is that located at the caudal end of the PSM and this has shown to be dependent on FGF signalling [34] , [35] . In principle , the expression of Hes7 in this region could be consistent with the existence of a Notch-independent pacemaker implicated in the initiation of the oscillations . However , the results produced after culturing half embryo explants from the same embryo for different periods of time in the presence of Notch-blocking drugs show that absence of Notch activity leads to a loss of dynamism in the expression of Hes7 and Axin2 in the caudal PSM , which indicates that at least in the mouse embryo there does not appear to be a Notch-independent pacemaker . It will be of great interest to study the mechanism by which Notch activity modifies the regulation of non-cyclic Hes7 and Axin2 in the caudal PSM such that it becomes oscillatory . As mentioned above , it is widely accepted that in the zebrafish embryo the main function of Notch is to synchronise the oscillations of her cyclic genes and that Notch inhibition does not interrupt the generation of oscillations and the resulting segmented body plan [7]–[11] , [62] . The present study does not provide evidence for a role for Notch in synchronizing the mouse somitogenesis oscillator , but it also does not seem to preclude such a role . Nevertheless our data clearly indicate that in mouse this signalling pathway plays a critical central role in the mechanism of the segmentation clock , which is not the case in zebrafish . This crucial difference in the role for Notch during mouse and zebrafish somitogenesis could be due to species-differences in the complexity of the core mechanism of the segmentation clock; an idea supported by the fact that in mouse the oscillator mechanism drives periodic expression of cyclic genes from three signalling pathways whereas in zebrafish the mechanism seems solely based on the oscillations of her genes which are both Notch and FGF dependent [2] , [36] . So far no Wnt-based cyclic genes have been reported in the zebrafish . In summary , our data show that in the mouse embryo Notch signalling is absolutely required to generate periodicity by the somitogenesis oscillator , as evidenced by the expression of cyclic genes and the formation of somites . The different signalling pathways implicated in this oscillator mechanism all appear to be interconnected via Notch signalling . A better knowledge of these reciprocal interactions will be of great relevance to gain a deeper understanding of the fundamental workings of this oscillatory mechanism .
Wild type CD1 Mus musculus embryos were obtained from timed mated pregnant females between 8 . 0 and 10 . 5 days postcoitum ( dpc ) . Lfng−/− , Hes7−/− , Rbpj−/− and Psen1−/−;Psen2−/− embryos were obtained and genotyped by PCR analysis of the yolk sacs as described [18] , [19] , [39] , [41] . For half embryos analysis E9 . 5–10 . 5 mouse embryos were isolated and the caudal portion was divided into two halves by bisecting the tissue along the midline . Explants were cultured on medium composed of DMEM/F12 supplemented with 10% fetal bovine serum , 10 ng/ml bFGF , and 50 U/ml penicillin/streptomycin . At the end of the culture period the explants were transferred into 4% paraformaldehyde fixative solution and then analysed by in situ hybridisation for gene expression . Four different series of experiments were performed: ( A ) One half explant was fixed and the other half was cultured for 60 minutes . ( B ) The two halves were cultured for 3 hours in medium in the absence or presence of either 100 µM DAPT ( Calbiochem ) or 100 nM LY411575 to inhibit the Notch activity , or the equivalent volume of DMSO as control . ( C ) The two halves were cultured for 3 hours in medium in the presence or absence of a mix of 100 µM DAPT plus 50 µM SU5402 ( Calbiochem ) . ( D ) The two halves were cultured for 3 hours in medium in the presence of 100 µM DAPT or 100 nM LY411575 , then one half was fixed and the second half was cultured for 1 additional hour before fixation . Whole embryo mouse culture was performed as previously described [63] , [64] using wild type embryos . In short , E8 . 0–8 . 5 mouse embryos with their membranes intact were cultured for about 18–20 hours in standard whole embryo roller culture conditions: 50% rat serum in F12 medium plus 1 mM sodium pyruvate , 2 mM glutamine and non-essential amino acids at 37°C with 5% CO2 . Media was supplemented with either 100 µM DAPT or 100 nM LY411575 or the appropriate amount of DMSO as control . All animals were handled in strict accordance with good animal practice as defined by the relevant national and/or local animal welfare bodies , and all animal work were approved by the ethical committees for experiments with animals of the University of Dundee ( UK ) , Nara Institute of Science and Technology ( Japan ) , University of Leuven ( Belgium ) and Centro Nacional de Investigaciones Cadiovasculares ( Spain ) . Mouse intronic and exonic Lfng , intronic and exonic Hes7 , Axin2 , Mesp2 , Snail1 , Uncx4 . 1 , Dusp6 and Sprouty2 probes were prepared as described [16] , [20] , [23] , [30] , [32] , [37] , [65] , [66] . Whole-mount in situ hybridisation was done basically as described [67] . The following modifications to this protocol were used for intronic probe in situ hybridisation . Samples were hybridised with probe for 40 hours in a low stringency hybridisation mix ( 50% formamide , 5× SSC , 5 mM EDTA , 50 µg/ml tRNA , 0 . 2% Tween-20 , 0 . 1% SDS , 100 µg/ml heparin ) and post-hybridisation washes were performed in post-hybridisation buffer ( 50% formamide , 0 . 1% Tween-20 , 1× SSC ) . Samples were processed either by hand or using the InsituPro VS Robot ( Intavis AG ) . All images were captured on a Leica MZ16 APO microscope using a Jenoptik camera . Images were recorded using Openlab software version 4 . 0 . 3 . The reaction was accomplished in the presence of SYBR Green Supermix ( BioRad ) and the reactions were measured in a Mastercycler ep realplex ( Eppendorf ) using the following cycling conditions: 95°C for 5 min , 40 cycles at 95°C for 15 sec and 53°C for 60 sec . Primers to quantify Hes7 mRNA levels were designed using Primer3 . The two primers used were 5′-GAAGCCGTTGGTGGAGAAG-3′ and 5′-GGCTTCGCTCCCTCAAGTAG-3′ . Normalization was performed against β-actin amplified using the primers 5′-GGCTGTATTCCCCTCCATCG-3′ and 5′-CCAGTTGGTAACAATGCCATGT-3′ . E18 . 5 mouse embryos were fixed in 95% ethanol overnight at room temperature and stained with 150 mg/ml Alcian Blue in 1∶4 mixture of acetic acid and 95% ethanol at RT for 24–48 h . After washing with 95% ethanol for 1 h , the embryos were treated with 1% KOH for 24 h with several changes . Embryos were subsequently stained with 75 mg/ml Alizarin Red S in 1% KOH solution for 12–24 h and cleared in a solution of 20% glycerol and 1% KOH for a week with daily changes . Samples were transferred to 50% glycerol , 50% ethanol for photography and storage . Whole-mount immunohistochemistry with anti-Hes7 and anti-NICD antibodies was performed as described previously [20] , [27] , [46] . Briefly , for analysis with the anti-Hes7 , embryos were fixed with 4% paraformaldehyde in PBS at 4°C for 3 h and treated with 0 . 1% H2O2 overnight . Then the embryos were incubated with anti-Hes7 antibody ( 1∶100 diluted ) at 4°C for 3–5 days and next with HRP-conjugated anti-guinea pig IgG ( Chemicon ) overnight at 4°C . The peroxidase deposits were visualized by 4-chloro-1-naphthol . For analysis with the anti-NICD on transversal sections 8-µm paraffin-embedded sections were immersed in 10 mM sodium citrate pH 6 . 0 and boiled 10 min to enable antigen retrieval . Immunostaining was performed with cleaved Notch1 antibody ( Val1744 , 1∶100 diluted , Cell Signaling Technology ) overnight at 4°C , followed by biotinylated anti-rabbit IgG antibody ( 1∶100 diluted , Vector Laboratories ) for 60 min at RT . Finally , the signal was amplified in two steps; first with avidin/biotin-HRP ( ABC kit , Vector Labs ) for 60 min at RT and then with Tyramide-Cyn3 ( NEL 744 , Perkin Elmer ) . Samples were prepared with caudal fragments of E9 . 5–10 . 5 mouse embryo control or treated with drugs and with mouse embryonic fibroblast derived from Psen1−/−;Psen2−/− embryos [47] . 20 µg of protein samples were used to perform electrophoresis in MOPS running buffer . Gels were then blotted and the resulting membrane was incubated with the anti-NICD antibody ( Val1744 , 1∶1000 dilution , Cell Signaling Technology ) overnight at 4°C followed by anti-rabbit-HRP antibody ( 1∶1000 ) for 60 min at RT and then standard ECL revelation ( Pierce ) . α-Tubulin staining was preformed with a 1∶20000 dilution ( Abcam , ab7291 ) followed by a 1∶2000 dilution of anti-mouse-HRP antibody . | Vertebrate animals generate their segmented body plan during embryogenesis . These embryonic segments , or somites , form one after another from tissue at the tail end of the embryo in a highly regulated process controlled by a molecular oscillator . This oscillator drives the expression of a group of genes in this tissue and determines the periodicity of somite formation . To date the genes regulated by this molecular clock comprise components of the Notch , Wnt , and FGF pathways . Recent data in the zebrafish embryo have suggested that the only role of Notch signalling in this process is to synchronise gene oscillations between neighbouring cells and that somite formation can continue in the absence of Notch activity . However , we show that mouse embryos lacking all Notch activity do not show oscillatory activity and do not develop somites . We propose that , at least in the mouse embryo , Notch activity is absolutely essential for building a segmented body axis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] | [
"developmental",
"biology/embryology",
"developmental",
"biology",
"developmental",
"biology/morphogenesis",
"and",
"cell",
"biology",
"developmental",
"biology/pattern",
"formation",
"developmental",
"biology/cell",
"differentiation",
"developmental",
"biology/molecular",
"develop... | 2009 | Notch Is a Critical Component of the Mouse Somitogenesis Oscillator and Is Essential for the Formation of the Somites |
Spatiotemporal pattern formation in neuronal networks depends on the interplay between cellular and network synchronization properties . The neuronal phase response curve ( PRC ) is an experimentally obtainable measure that characterizes the cellular response to small perturbations , and can serve as an indicator of cellular propensity for synchronization . Two broad classes of PRCs have been identified for neurons: Type I , in which small excitatory perturbations induce only advances in firing , and Type II , in which small excitatory perturbations can induce both advances and delays in firing . Interestingly , neuronal PRCs are usually attenuated with increased spiking frequency , and Type II PRCs typically exhibit a greater attenuation of the phase delay region than of the phase advance region . We found that this phenomenon arises from an interplay between the time constants of active ionic currents and the interspike interval . As a result , excitatory networks consisting of neurons with Type I PRCs responded very differently to frequency modulation compared to excitatory networks composed of neurons with Type II PRCs . Specifically , increased frequency induced a sharp decrease in synchrony of networks of Type II neurons , while frequency increases only minimally affected synchrony in networks of Type I neurons . These results are demonstrated in networks in which both types of neurons were modeled generically with the Morris-Lecar model , as well as in networks consisting of Hodgkin-Huxley-based model cortical pyramidal cells in which simulated effects of acetylcholine changed PRC type . These results are robust to different network structures , synaptic strengths and modes of driving neuronal activity , and they indicate that Type I and Type II excitatory networks may display two distinct modes of processing information .
Neuronal synchronization is thought to underlie spatiotemporal pattern formation in the healthy [1]–[4] and pathological brain [5]–[9] . The propensity for synchronization in a neuronal network is determined by both cellular and network properties . An important experimentally obtainable measure of cellular properties is the neuronal phase response curve ( PRC ) [10] . The PRC characterizes the change in spike timing of a periodically firing neuron in response to brief , weak external stimulation . PRCs have been classified into two general categories: Type I , which display only phase advances in response to excitatory stimuli , and Type II , which respond with both phase advances and delays . Type I cells exhibit relatively poor propensity for synchronization under excitatory coupling , while Type II cells synchronize better [10]–[17] . Furthermore , the PRC characteristics thought to be responsible for synchronization propensity change differentially as a function of frequency for Type I and Type II cells [18] . In this study , we explain the differential effects of frequency modulation on neuronal response properties and exploit these effects to investigate differential changes in the capacity for synchronization of excitatory networks consisting of Type I or Type II neurons . To demonstrate the universality of the frequency-dependent effects on the neuronal PRC , we consider a reduced model neuron described by the Morris-Lecar equations [19] which can display either a Type I or Type II PRC in different parameter regimes [20] . Then , to present the effects within a physiological context , we turn to the results of a recent experimental study which showed that cholinergic modulation of cortical pyramidal neurons switches the neuronal PRC from Type II to Type I [21] . In a Hodgkin-Huxley-based cortical pyramidal neuron model , the switch in PRC type was shown to depend on a slow , low-threshold potassium current which is targeted by cholinergic modulation [22] . Using these two neuronal models , we explain the underlying cellular basis of the differential frequency effects on the PRC . We show that the relative timing of hyperpolarizing , potassium currents in relation to the model's depolarizing currents ( a calcium current in the Morris-Lecar model and a sodium current in the cortical pyramidal cell model ) plays a crucial role in shaping the phase response of a neuron . We then investigate the influence of the frequency-dependent cellular effects on network activity by analyzing network synchronization as a function of underlying neuronal spike frequency near firing threshold in large-scale , excitatory networks composed of either Morris-Lecar neurons or cortical pyramidal model neurons . As expected , the neuronal PRC type profoundly affects network propensity for synchronization [17] . We show that , in general , increasing firing frequency near firing threshold has little effect upon synchrony in Type I networks , while it severely suppresses synchrony in Type II networks . We show these results to be robust to neuronal heterogeneity , network connectivity parameters and whether neuronal activity is driven by constant or stochastic inputs . Our results provide important insight into differential changes in the propensity for network synchronization induced by the external modulation of neuronal frequency . As neuronal firing frequency changes , the changes in network spatiotemporal patterns depend upon the response characteristics of the individual cells in the network .
We used the Morris-Lecar model [19] as a generic neuronal model to initially explore frequency-dependent PRC effects . The model contains two active ionic currents: an inward current whose dynamics are instantaneous and an outward current gated by the dynamic variable . The current balance equation for the cell is ( 1 ) where , is in millivolts , is in milliseconds , is an externally applied current measured in , and is the synaptic current received by neuron . The current is governed by the steady-state activation function , while dynamics of the current gating variable are given by , with and . The Type I and Type II neuronal models share the parameter values , , , , , , and . Type I cells are modeled with , , , and , while Type II cells are modeled with , , , and . These values were taken from [20] . The cortical pyramidal model neuron we employed was motivated by recent computational and experimental findings , as reported in [22] . Varying the maximum conductance of a -mediated adaptation current , , from to effectively switches the response characteristics of the cortical pyramidal model neuron from Type II to Type I , a phenomenon which has been observed in situ and simulates the effects of cholinergic neuromodulation [21] . The model also features a fast , inward current , a delayed rectifier current , and a leakage current , in addition to the aforementioned slow , low-threshold current responsible for spike-frequency adaptation [22] , [23] . The current balance equation for the cell is ( 2 ) with , in millivolts , and in milliseconds . is an externally applied current measured in , and is the synaptic current received by neuron . Activation of the current is instantaneous and governed by the steady-state activation function . Dynamics of the current inactivation gating variable are given by ( 3 ) with and . The delayed rectifier current is gated by , whose dynamics are governed by ( 4 ) with and . The slow , low-threshold current targeted by cholinergic modulation is gated by , which varies in time according to ( 5 ) where . The parameters and in the current gating equations are varied in the investigation of the underlying cellular basis of the differential frequency effects on the PRC , but they are set to in the network simulations . The slow , low-threshold current loosely models the muscarine-sensitive M-current observed in cortical neurons . It has been shown in silico that eliminating this current is sufficient to switch the model neuron's PRC from Type II to Type I [22] . This is intended to model cholinergic neuromodulation , which has been shown experimentally to switch cortical pyramidal neurons between Type I and Type II phase responses [21] . This switching of PRC profile is demonstrated in Fig . 1 , and is obtained by setting to obtain a Type I response ( simulated cholinergic modulation ) and to obtain a Type II response ( simulated absence of cholinergic modulation ) . All other parameter values are the same for both types of neurons: , , , , , and . For both neuronal models , is set to a fixed value to elicit repetitive firing in a single , synaptically isolated neuron , and the model equations are time evolved using a fourth-order Runge-Kutta numerical scheme until the oscillatory period stabilizes . Then , using initial conditions associated with spike peak , brief current pulses are administered at different phases of the oscillation , and the perturbed periods are used to calculate the corresponding phase shifts . The current pulses are administered at 100 equally-spaced time points throughout the period of the neuronal oscillation . The current pulses have a duration of 0 . 06 ms and an amplitude of 3 . 0 for the Type I cortical pyramidal neuron , a duration of 0 . 06 ms and an amplitude of 10 . 0 for the Type II cortical pyramidal neuron , and a duration of 0 . 50 ms and an amplitude of 100 . 0 for both the Type I and Type II Morris-Lecar neurons . In all network simulations , the number of neurons is 200 , and the synapses are exclusively excitatory . The network connectivity pattern is constructed using the Watts-Strogatz architecture for Small World Networks [24] . Starting with a 1-D ring network with periodic boundary conditions , each neuron is at first directionally coupled to its nearest neighbors , and then every connection in the network is rewired with probability to another neuron selected at random . In this way , results in a locally-connected network and in a randomly connected network . The radius of connectivity therefore determines the density of connections in the network , while the re-wiring parameter determines the network connectivity structure . Network connectivity is set to 4 in all simulations . Synaptic current is transmitted from neuron at times when its membrane voltage breaches −20 mV . The synaptic current delivered from neuron to a synaptically connected neuron at times is given by . The total synaptic current to a neuron is simply given by , where is the set of all neurons which synapse onto neuron . The synaptic weight s is the same for all synapses within a given simulation , and we set and . All simulations are run for 10 , 000 ms , with the first 3000 ms disregarded in order to eliminate initial transient effects . The dynamics are numerically integrated in Matlab using a fourth-order Runge-Kutta method with a time step of 0 . 05 ms for the cortical pyramidal neuron networks and 0 . 10 ms for Morris-Lecar neuron networks . We employ two different methods to modulate network firing frequency in our simulations . The first is to simply modulate the supra-threshold value of for all neurons in the network . In order to prevent the networks from trivially synchronizing , we do not supply each neuron with exactly the same level of current , but instead sample from a Gaussian distribution of current values . The mean value of the distribution determines the average firing frequency of the network , and the standard deviation of the Gaussian is chosen such that the standard deviation in natural neuronal frequencies is 1 Hz . In order to model more biologically relevant environmental inputs , we also run simulations of cortical pyramidal neuronal networks in which frequency is modulated by stochastic input . All neurons are given the same constant sub-threshold baseline current , plus square current pulses randomly delivered to each neuron at a specified frequency , so that . The delivery of the square current pulses is a Poisson process . Modulation of this noise frequency thereby modulates the average frequency of the network . In our simulations of stochastically-driven cortical pyramidal neuronal networks , consists of square current pulses with amplitude 30 and duration 0 . 2 msec . With these values , at least two successive pulses are required to elicit neuronal firing . The baseline currents are for Type I networks and for Type II networks . We monitor phase-synchronization of neuronal firing in our simulations using the mean phase coherence ( MPC ) measure , [25] . This measure quantifies the degree of phase locking between neurons , assuming a value of 0 for completely random spiking and 1 for complete phase locking . Note that MPC may be attained for locking of phases at any value , not just zero . The MPC between a pair of neurons , , is defined by: ( 6 ) ( 7 ) where is the time of the spike of neuron 2 , is the time of the spike of neuron 1 that is largest while being less than , is the time of the spike of neuron 1 that is smallest while being greater than or equal to , and is the number of spikes of neuron 2 . The MPC of the entire network , , is calculated by averaging over all pairs of neurons , excluding . Note that this measure is not symmetric . We quantify phase-zero synchronization of a network by calculating the bursting measure B , which is 0 for random spiking and approaches 1 for perfect locking at phase zero between all neurons , for a large number of total spikes and neurons . Calculation of B requires a time-ordered list of the spike times of all neurons over the duration of the entire simulation [26] . Denoting as the time difference between spikes i and i+1 , which do not necessarily ( and probably do not ) correspond to spikes of the same neuron , B is then defined as ( 8 ) where represents averaging over all spikes . This measure makes use of the fact that an ensemble of spike time intervals will have a larger standard deviation in a synchronous signal than in an asynchronous signal . In our simulations , both the mean phase coherence and the bursting parameter are calculated for neuronal activity from 3000 ms to 10 , 000 ms , unless otherwise noted .
Fig . 1 displays the response properties of the model neurons in our simulations , with Fig . 1A , D showing the frequency-current curves of the model neurons and Fig . 1B , C , E , F showing the PRCs of the model neurons . Type I PRCs in both the Morris-Lecar and the cortical pyramidal neuron models exclusively display phase advances ( positive PRC values ) in response to excitatory perturbations ( Fig . 1B , E ) while Type II PRCs show phase delays ( negative PRC values ) at earlier phases and advances at later phases ( Fig . 1C , F ) . ( Note that the presence of small negative regions early in Morris-Lecar PRCs and the absence of such regions in cortical pyramidal cell PRCs is a consequence of the fact that spikes consume a much larger portion of the interspike interval in the Morris-Lecar model than in the cortical pyramidal cell model [20] . We therefore ignore these early regions in Morris-Lecar PRCs . ) The switch from Type I to II is induced by changes in the steady state activation function of the current in the Morris-Lecar model and by the presence of the slow , low-threshold current in the cortical pyramidal cell model . A categorization of Type I and Type II can also be applied to a neuron's frequency-current ( f-I ) relation , with Type I f-I curves exhibiting arbitrarily low frequencies at firing thresholds and Type II f-I curves showing a finite , non-zero firing frequency at threshold . While the categorization of a neuron's PRC and f-I curve are not necessarily the same , and the relationship between the curves has not been completely determined [27] , for both models considered here , PRC and f-I curve types coincide ( Fig . 1A , D ) . In both models , increasing frequency by increasing the constant applied current results in an attenutation of phase responses ( Fig . 1 ) . This attenuation occurred in qualitatively different ways for Type I and Type II neurons . In the Type I model neurons , increased firing frequency led to diminished phase advances but did not change the relative shape of the curves–they all remained distinctly Type I ( Fig . 1B , E ) . In the Type II model neurons , however , there was much greater attenuation of the phase-delay region compared to the phase-advance region ( Fig . 1C , F ) . This asymmetric attenuation can affect synchronization properties because the phase-delay region contributes to the increased propensity for synchronization in Type II excitatory networks [12] . Previously , the emergence of phase delay regions at low firing frequencies was attributed to decreased activation of -mediated adaptation currents at low frequencies [12] , [18] , but this explanation cannot apply to the Morris-Lecar model , since it contains no adaptation currents . Below we discuss the properties of a cell's hyperpolarizing and depolarizing currents that are responsible for its phase response , and which explain the observed frequency-dependent attenuation . In both models , phase delays exist in the Type II parameter regimes because there is a voltage interval in which activation of an outward , hyperpolarizing current is greater than activation of the inward , depolarizing current . In the Type II Morris-Lecar model , the steady state activation curve of the current , , is shifted to the left and steeper compared to that of the current , , thus providing for this voltage interval . In the Type II cortical pyramidal neuron model , the steady state activation curve , , of the slow , low-threshold current ( which is absent in the Type I neuron ) , is similarly shifted to the left relative to the steady-state activation curve of the current , . In either model , as the voltage trajectory passes through the early part of the interspike interval , a brief , excitatory stimulus will induce a larger response from the lower-threshold current than from the inward current , resulting in negative values of the PRC at early phases . At higher voltage levels later in the interspike interval , the inward current dominates the response to the brief stimulus due to its faster ( instantaneous ) activation dynamics , thus leading to advances in the cycle , and positive values of the PRC at later phases . As firing frequency increases , the cycle trajectory passes through this -dominant voltage interval at a faster rate , thus preventing the full response from developing before reaching voltage levels where the instantaneous inward current can respond . The delaying response to the brief stimulus is thus diluted by the advance-promoting inward current response , and phase delays are attenuated . This attenuation of phase delays is therefore the result of a disparity between the fixed dynamics of the delay-inducing current and the time afforded that current to act by the shrinking interspike interval . Phase advances are less sensitive to frequency modulation since the instantaneous dynamics of the inward currents in both models can directly track the faster cycle trajectory . We further illustrate this point by modulating the speed of the gating variable controlling the delay-inducing potassium current in each model . Fig . 2A demonstrates that in the Morris-Lecar model , increasing , which increases the rate of the gating variable , results in an increase in the amplitude of PRC phase delays , while decreasing has the opposite effect . Faster dynamics allow for faster development of the delaying response to the excitatory stimulus . In this model , modulating also changes the voltage levels during the interspike interval , which can shift the dominant voltage interval to different phases . We systematically quantify the contribution of dynamics to the generation of the phase delay by measuring the changes in the PRC delay depth as a function of for neurons receiving different driving currents and thus exhibiting different intrinsic frequencies ( Fig . 2B ) . The depth of the PRC delay region increased with increasing for all levels of external current , and faster-firing neurons could display similar delay depths as slower-firing neurons with appropriate increases in . While increasing also acted to increase firing frequency ( Fig . 2C ) , phase delay amplitudes nonetheless increased , indicating that speeding up the rate of dynamics exerts a stronger effect on the phase delay than does the accompanying frequency increase . A similar dependence of phase delay amplitude on the rate of the gating variable for the slow , low-threshold current in the cortical pyramidal cell model is shown in Fig . 2D , E . As the rate of dynamics increased ( i . e . , as increased in Eq . 5 ) , depths of the PRC delay increased due to the ability of the current to develop a delaying response before voltage levels were reached where the current activated . Again , increasing the rate of dynamics caused an increase in frequency ( Fig . 2F ) , but the faster development of the response to the perturbation could overcome a frequency-induced attenuation of phase delays . In this model , voltage levels during the interspike interval also changed with the changes in , but they did not greatly influence the phase of maximal delays . In the cortical pyramidal neuron model , the amplitude of phase delays also depended on the rate of the current inactivation ( Fig . 2G–I ) , gated by the variable in Eq . 2 . Slower inactivation , induced by lower values of in Eq . 3 , allowed larger responses to the perturbing stimulus , which diluted the delaying effect of the response and therefore attenuated phase delays . The rate of inactivation had little effect on voltage levels as a function of phase during the interspike interval , and only slightly affected the frequency . Increasing the rate of inactivation did induce a decrease in firing frequency , which would promote the observed increase in delay depth , but these changes to firing frequency were too slight to be the primary cause of the enlarged delay amplitude . These results imply that appropriate selection of the rate of variables gating the intracellular currents mentioned above permits the recovery of specified PRC delay depths for different levels of external current . Fig . 3 illustrates this effect for both models . From the curves in Fig . 2B , E , H , appropriate rates of the gating variables were separately selected for each level of external current to induce delay depths of 0 . 04 in the Morris-Lecar neuron and 0 . 025 in the cortical pyramidal neuron . In the Morris-Lecar model , the maximal phase delay region was shifted to the left as the external current increased because the voltage trace was similarly shifted ( Fig . 3A , D ) . However , in the cortical pyramidal cell model , the PRC profiles were virtually identical for different levels of external current , both when the slow potassium current was modified and when the sodium inactivation was modified ( Fig . 3B , C ) . This was due to the fact that the voltage traces ( plotted as a function of oscillatory phase ) were not shifted when either of these intracellular currents were altered ( Fig . 3E , F ) . The invariance of the voltage traces in the cortical pyramidal cell model is an interesting phenomenon , but it is beyond the scope of this paper .
We have shown that excitatory networks composed of neurons with either Type I or Type II PRC properties respond very differently to frequency modulation near firing threshold , with Type I network synchrony remaining largely unaffected by frequency modulation and Type II networks synchronizing much better at lower frequencies . This result is robust in virtually all network parameter regimes in which the network is capable of attaining any appreciable level of synchronization . While both Type I and Type II PRCs are modified by changes in frequency , only Type II PRCs change in qualitative profile . Specifically , the phase delay region , which is known to be critical in promoting synchrony , is severely attenuated . Increased frequency therefore tends to have little effect upon Type I networks , since there is no change in the PRC's contribution to synchrony , while in Type II networks it leads to depressed synchrony via the diminished phase delay region of the PRC . It should be noted that our simulations agreed with a large body of previous work showing that neurons with Type II membrane dynamics ( as defined by the frequency-current curve ) tend to synchronize better than neurons with Type I membrane dynamics , when coupled with excitation . Previous theoretical work indicates that when excitatory networks are driven with constant current , those composed of Type I neurons will not synchronize as well as those composed of Type II neurons [10] , , a phenomenon which we observed in our simulations , since much larger synaptic coupling values were required in Type I networks to evoke levels of synchrony equivalent to those in Type II networks ( Fig . 6 ) . Previous theoretical [29] and experimental [30] work has also shown that neurons with Type I membrane dynamics respond to excitatory noisy input with much higher spike-time variability than do neurons with Type II membrane dynamics . This accords with the results of our simulations of networks stimulated by noisy current pulses , where again we saw that greater synaptic coupling was needed for Type I networks to synchronize as well as Type II networks ( Fig . 10 ) . In this study , we focused on the implications for network synchronization of the observed frequency-dependence of PRCs . Our results suggest that the severe attenuation of the phase-delay region of Type II PRCs at increased firing frequencies contributes to the observed decline in network synchronization at such frequencies . Frequency-dependent modification of PRCs has been investigated before in complex , multi-compartment neuronal models [31] , [32] , but such results rely on dendritic effects and hence do not apply to our results using single-compartment neurons . It has been shown in a simple -neuron model that low-threshold adaptation currents can produce negative regions in the PRC at low frequency [18] , an effect which is probably due to the change in bifurcation structure induced by such currents [12] . From this perspective , the delay region of the PRC develops only at low frequency because the adaptation current is saturated at high frequency , resulting in its responding to excitatory stimulation with relatively smaller transient increases . Our work extends this insight by explaining the emergence and attenuation of delay regions in the PRCs of Morris-Lecar neurons , which have no adapting current . Our explanation applies to the cortical pyramidal model neuron , which does feature an adapting current , as well: it is the speed of low-threshold , hyperpolarizing currents relative to the interspike interval which determines the depth of the PRC delay region in Type II cells . For a fixed level of external current , the faster we made the current in the Morris-Lecar neuron and the adapting current in the cortical pyramidal neuron , the larger their PRC delay depths grew . It is therefore not only the saturation level of low-threshold , hyperpolarizing currents that is important , but also the speed with which they can respond to brief stimulation . In addition , our simulations showed that the PRC delay depth is not exclusively controlled by the effects of hyperpolarizing currents , but can be greatly affected by depolarizing currents as well . The faster we made the deactivation of the sodium current , the larger the delay depth grew , underscoring once again the importance of the speed of intracellular currents relative to the interspike interval . The frequency-dependent synchronization which we have described in this paper could potentially be involved in any cognitive process , functional or pathological , which involves spatiotemporal pattern formation of neuronal populations . For example , cholinergically-induced switching between sensitivity and insensitivity to frequency modulation could be important in proper memory consolidation during slow wave and REM sleep , two states that are characterized by differing levels of acetylcholine in cortical and hippocampal regions . Frequency-mediated synchrony could also play a part in the binding of signals from multiple sensory modalities . Gamma oscillations ( 20–80 Hz ) in cortical networks are believed to be generated by synchronous activity of fast-spiking interneurons [33] , which generally exhibit Type II frequency-current relations and PRC profiles [27] , [30] . While excitatory and inhibitory synaptic connections and gap junctions may participate in the synchronous firing of interneuron networks [34]–[37] , our results suggest the importance of the cellular properties of the fast-spiking interneurons in generating synchrony . Additionally , the frequency-dependence of synchronization may provide a means to restrict synchronization to specific frequency bands . Finally , frequency modulation could contribute to the onset of epileptiform activity , and our results might help to explain recent evidence that synchrony decreases during seizures [38] , [39] . At the same time , the importance of our results is not confined to these examples alone . Our findings point to the possibility that Type I and Type II excitatory networks function in two separate coding regimes , with Type I networks functioning in the rate coding regime and Type II networks functioning in the temporal coding regime , effectively acting as low-pass filters . Further experimental investigation into the interplay between cellular properties , frequency , and network synchronization is clearly required . | Synchronization of the firing of neurons in the brain is related to many cognitive functions , such as recognizing faces , discriminating odors , and coordinating movement . It is therefore important to understand what properties of neuronal networks promote synchrony of neural firing . One measure that is often used to determine the contribution of individual neurons to network synchrony is called the phase response curve ( PRC ) . PRCs describe how the timing of neuronal firing changes depending on when input , such as a synaptic signal , is received by the neuron . A characteristic of PRCs that has previously not been well understood is that they change dramatically as the neuron's firing frequency is modulated . This effect carries potential significance , since cognitive functions are often associated with specific frequencies of network activity in the brain . We showed computationally that the frequency dependence of PRCs can be explained by the relative timing of ionic membrane currents with respect to the time between spike firings . Our simulations also showed that the frequency dependence of neuronal PRCs leads to frequency-dependent changes in network synchronization that can be different for different neuron types . These results further our understanding of how synchronization is generated in the brain to support various cognitive functions . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"biology",
"computational",
"biology",
"neuroscience"
] | 2011 | Cellularly-Driven Differences in Network Synchronization Propensity Are Differentially Modulated by Firing Frequency |
Directed cell migration in response to chemical cues , also known as chemotaxis , is an important physiological process involved in wound healing , foraging , and the immune response . Cell migration requires the simultaneous formation of actin polymers at the leading edge and actomyosin complexes at the sides and back of the cell . An unresolved question in eukaryotic chemotaxis is how the same chemoattractant signal determines both the cell's front and back . Recent experimental studies have begun to reveal the biochemical mechanisms necessary for this polarized cellular response . We propose a mathematical model of neutrophil gradient sensing and polarization based on experimentally characterized biochemical mechanisms . The model demonstrates that the known dynamics for Rho GTPase and phosphatidylinositol-3-kinase ( PI3K ) activation are sufficient for both gradient sensing and polarization . In particular , the model demonstrates that these mechanisms can correctly localize the “front” and “rear” pathways in response to both uniform concentrations and gradients of chemical attractants , including in actin-inhibited cells . Furthermore , the model predictions are robust to the values of many parameters . A key result of the model is the proposed coincidence circuit involving PI3K and Ras that obviates the need for the “global inhibitors” proposed , though never experimentally verified , in many previous mathematical models of eukaryotic chemotaxis . Finally , experiments are proposed to ( in ) validate this model and further our understanding of neutrophil chemotaxis .
Chemotaxis , the directed movement of cells in response to chemical gradients , plays a prominent role in a number of physiological processes , including foraging , wound healing , tumor metastasis , and the immune response [1–3] . In the case of the immune response , chemoattractants are produced at or proximal to sites of infection . Leukocytes sense these chemoattractants and move in the direction where the chemoattractant concentration is greatest , thereby locating infected tissue and invading microbes . Leading this assault are neutrophils . These cells circulate in the bloodstream and upon activation squeeze through the vascular endothelium and crawl to sites of infections and inflammation . There they phagocytose bacteria and release a number of proteases and reactive oxygen intermediates with antimicrobial activity [4 , 5] . Neutrophils sense chemoattractants using transmembrane receptors , primarily G protein–coupled receptors ( GPCR ) , which are evenly distributed along the plasma membrane [6] . Binding of chemoattractants to these receptors activates a complex network of interacting proteins , lipids , and small molecules . This signaling cascade leads to a symmetry-breaking event , where a number of regulatory proteins and lipids , initially distributed uniformly on either the membrane or in the cytosol , are recruited to either the front or back of the cell upon stimulation with attractant . Differential localization of these proteins serves as a compass for the migrating cell [7 , 8] . A number of experimental studies have identified and characterized the core regulators necessary for eukaryotic chemotaxis using the slime mold Dictyostelium discoideum , primary neutrophils , and HL-60 cells , which are a myeloid leukemia that can be differentiated into a neutrophil-like state [9 , 10] . The basic model for neutrophils is as follows . When neutrophils or differentiated HL-60 cells are stimulated with ligand , typically formyl-methionylleucylphenylalanine ( fMLP ) , two divergent pathways are activated [11] . In the first , or “frontness” pathway , receptor-activated Gi proteins activate the membrane-bound G-protein Ras and recruit phosphatidylinositol-3-kinase ( PI3K ) to the membrane at the leading edge of the migrating cell [12] . Activated Ras subsequently binds to membrane-bound PI3K , yielding a complex that begins the conversion of phosphatidyl-4-5-bisphophate ( PIP2 ) to phosphatidyl-3-4-5-triphosphate ( PIP3 ) [13–15] . PIP3 then recruits the Rho-GTPases Rac and Cdc42 to the leading edge of the membrane [16 , 17] . Rac and Cdc42 associate with the WASP/SCAR complex that stimulates actin polymerization . PIP3 , Rac , and actin also further stimulate the conversion of PIP2 to PIP3 via a still uncharacterized positive-feedback mechanism [16 , 18–20] . In the second , or “backness” pathway , receptor-activated G12|13 proteins activate and then recruit the Rho GTPase RhoA to the membrane at the lagging edge of the migrating cell [11 , 21 , 22] . RhoA plays a role in activating myosin contractions [11] . It also activates and/or recruits to the membrane the phosphatidylinositol phosphatase ( PTEN ) , which antagonizes the action of PI3K by converting PIP3 to PIP2 [21] . A number of mathematical models have been proposed to explore different potential mechanisms for gradient sensing and spatial localization in eukaryotic chemotaxis . A common mechanism in many of these models is the interplay between a local activator and a global inhibitor [23–27] . The activator binds to the membrane at a rate proportional to the local degree of receptor activation . Hence , more activator is bound at the front than at the rear of the cell in relation to the source of the chemoattractant . The inhibitor , on the other hand , responds to the integrated receptor activity . Its activity , therefore , is proportional to the average concentration of attractant across the length of the cell . Typically , the global inhibitor is assumed to be a rapidly diffusing protein or small molecule either on the membrane or in the cytosol . The cell determines its front and rear by comparing the local concentration of the activator on the membrane relative to the global concentration of the inhibitor . At the front , the concentration of the activator is greater than the inhibitor and vice versa at the rear . The popularity of the local activator/global inhibitor model is that it provides a simple mechanism for explaining how a cell can distinguish its front and rear from a common signal [28] . While the basic local activator/global inhibitor models correctly account for localization of proteins to the leading edge of the cell , these models fail to account for localization of associated proteins to the lagging edge . Ma and colleagues addressed this problem by proposing two opposing local activator/global inhibitor mechanisms; one for the cell's front and the other for the rear [29] . Alternatively , Narang proposed a model with two mutually exclusive local activators and a single global inhibitor [30] . One activator is directed to the leading edge by the standard local activator/global inhibitor mechanism , and the other is forced to the rear of the cell due to exclusion by the first at the front . Both models are able to selectively localize proteins to either the front or rear of the cell in response to a chemical gradient . However , these models include mechanisms involving global inhibitors , something that has not yet been experimentally corroborated . A number of alternative models for eukaryotic chemotaxis not involving global inhibitors have also been proposed . Postma and van Haastert proposed a simple mechanism for gradient sensing involving local activation coupled with positive feedback and substrate depletion of a single diffusing second messenger [31] . A similar mechanism tailored to fibroblasts was later proposed by Schneider and Haugh , where the unknown second messenger was replaced by PI3K [32] . Lacking in these two models , however , is a mechanism to explain the dynamics at the rear of the cell . To address this problem , Skupskey and colleagues proposed a model consisting of positive feedback , exact adaptation , and inhibition of PTEN by PI3K [33] . Likewise , Gamba and colleagues proposed a simple mechanism for polarization based solely on the antagonizing action of PI3K and PTEN and a positive feedback loop involving PTEN [34] . Meier-Schellersheim and colleagues , on the other hand , proposed a detailed model for gradient sensing and polarization involving a number of additional regulatory proteins such as Src and Paxillin [35] . In the process , they were able to explain interesting dynamics in Dictyostelium chemotaxis . While these models are able to explain many aspects of eukaryotic chemotaxis , they fail to address a number of mechanisms and behaviors specifically associated with neutrophils , in particular the role of the actin cytoskeleton in regulating chemotaxis . Here we propose a mathematical model for neutrophil gradient sensing and spontaneous polarization that does not require a global inhibitor . In this model , polarization of the front and back molecules is achieved by the switch-like activation of a coincidence circuit that requires both Ras and PI3K to transmit a signal . Our model is based on a phase-separating circuit and can reproduce many experimental data , including the effect of F-actin inhibitors . The model also exhibits partial adaptation when exposed to uniform concentrations of chemoattractant and forms signaling patches when these levels fluctuate as observed in a number of experiments [18 , 36 , 37] .
Figure 1 shows a schematic diagram of our proposed model of neutrophil directional sensing and polarization . The model is based on the general qualitative model proposed by Bourne and colleagues [11] . Briefly , the model assumes that ligand-bound receptors activate two pathways in parallel . In the first or “frontness” pathway , ligand-bound receptors activate Ras and recruit inactive PI3K to the membrane . Activated Ras then binds to membrane-bound PI3K , and this complex begins the conversion of PIP2 to PIP3 . PIP3 then stimulates actin polymerization and further enhances the activity of the Ras–PI3K complex . The second reaction is used to model the PIP3 positive feedback loop . A number of additional proteins including Rac , Cdc42 , and WASP are also known to be involved in the “frontness” pathway , but were not accounted for explicitly in the model . We chose to omit these proteins since their unique roles are still unknown and instead lumped them into the positive feedback loop involving PIP3 . In the second , or “backness” pathway , ligand-bound receptors activate and recruit RhoA to the membrane at the rear of the migrating cell . RhoA activates myosin contractions and is proposed to activate cytosolic PTEN . Once activated , PTEN binds to the membrane and converts PIP3 to PIP2 . The two parallel pathways in our model cross-inhibit each other at five junctions: ( 1 ) myosin contraction inhibits F-actin polymerization and vice versa; ( 2 ) F-actin causes RhoA to disassociate from the membrane; ( 3 ) F-actin causes PTEN to disassociate from the membrane; ( 4 ) myosin inhibits the formation of the PI3K–Ras complex; and ( 5 ) PTEN converts PIP3 to PIP2 while PI3K has reciprocal activity . These cross-inhibitory reactions serve to differentially localize and then stabilize these two pathways to the respective front and rear of the migrating cell . In our model formulation , the cell is treated as a disk in two dimensions , and the spatial positions of the different proteins and molecules on the membrane are represented using a single variable θ , taking values between zero and one . Figures 2–14 show plots of the concentrations of the various proteins and molecules as a function of membrane position θ and time . The interior of the cell is assumed to be well-mixed , and the cytosolic proteins are assumed to be spatially uniform . To limit the complexity of the plots , dynamic changes in the concentrations of the cytosolic proteins are omitted . The mathematical details and governing assumptions of the model are described in the Materials and Methods section , and Matlab simulation scripts are available at http://openwetware . org/wiki/Rao_Lab:Code . We first demonstrate that the model can correctly polarize and stabilize the respective components of the “frontness” and “backness” pathways when exposed to a gradient and still be responsive to changes in the gradient direction . The model was first simulated with a linear 50% gradient of chemoattractant ( that is , there was 50% more attractant at the front than the back ) , and then the gradient was slowly rotated by 180° . Figure 2 shows the spatial dynamics of key components of the model where the abscissa is time ( dimensionless units ) , the ordinate ( θ ) is the position on the membrane , and the color intensity is the normalized concentrations of the different components of the model . As documented in Figure 2 , both the “frontness” and “backness” components correctly polarize in response to a gradient . In particular , components of the “frontness” pathway localize to regions of the membrane where the ligand concentration is greatest , whereas components of the “backness” pathway localize to regions where the ligand concentration is weakest . Localization of the two pathways is also mutually exclusive; the respective components do not localize to the same regions of membrane . Furthermore , the cell is able to correctly repolarize after a change in the gradient direction . This result demonstrates that our model does not exhibit “lock-on” behavior , where the cell will first correctly polarize in response to a gradient but then be unable to change the direction of its polarity in response to a change in the direction of the gradient [23] . Note that there is a lag in the cell's response to a moving gradient ( Figure 2 ) . In other words , the cell is predicted to perform a U-turn rather than spontaneously repolarize in response to a change in the direction of the gradient . The lag is due to the dynamics of protein dissociation from the membrane and actin/myosin disassembly . If the gradient is rotated instantaneously , then the cell does not respond and exhibits “lock-on” behavior . The reason is that there is no initial cue to determine whether the cell should perform either a right or left U-turn , and , as a result , no choice is made . Consistent with this hypothesis , the cell is able to respond when the gradient is instantaneously rotated by 179° or less . Likewise , if rotation is fast but not instantaneous ( 0 . 01 time units ) , then the cell correctly polarizes in the direction of the new gradient ( Figure 3A ) ; however , the lag is larger . On the other hand , if the gradient is rotated at a slower rate ( 80 versus 40 time units ) , then the lag between the direction of the gradient and response is reduced ( Figure 3B ) because the speed of rotation matches the internal dynamics of the pathway . Polarization is known to be the result of the switch-like activation of PI3K and exclusion of RhoA by actin polymerization at the front of the cell [22] . In our model , we propose that PI3K activation depends on both the activation of Ras and the recruitment of cytosolic PI3K to the membrane . Class IB PI3Ks have localization ( p101 regulatory subunit ) and Ras binding domains , and it has been proposed that both are important regulators of PI3K signaling [12 , 15 , 38] . In this way , PI3K activation behaves like a coincidence circuit: both of its precursors , active Gβγ and Ras , must be present at a particular position on the membrane for activation to occur . This circuit gives a sharp peak of active PI3K ( i . e . , Ras–PI3K complex ) near receptors with the highest degree of ligand occupancy . The net result is that activation of PI3K is more sensitive than RhoA to the spatial signal generated by the chemoattractant gradient . As a consequence , PI3K activity is concentrated at the leading edge of the migrating cell and RhoA is forced to the rear due to inhibition by actin polymerization , a downstream response of activated PI3K . This asymmetry is amplified by the positive feedback loop involving PIP3 and also by the antagonizing action of F-actin and myosin . These results are illustrated in Figure 4 , which shows the steady-state distribution of receptor–ligand complexes ( black ) , activated PI3K ( Ras–PI3K ) ( red ) , and RhoA ( blue ) in response to a 50% linear gradient . The dynamic response is shown in Video S1 . We next considered how sensitive the model was to different gradient strengths and background concentrations . The following function was used to specify the ligand concentration along the surface of the cell where bL denotes the background concentration ( specified at the midpoint of the cell , θ = 0 . 5 ) and gL denotes the gradient strength ( given in terms of percentage ) . This function describes the surface concentration of a circular cell in a linear gradient . We first fixed the background concentration at bL = 1 and then varied the gradient strength from 10−3% to 104% . This background concentration matches the dimensionless KD for receptor–ligand binding in the model . As illustrated in Figure 5A , the model predicts that neutrophils are able to polarize in gradients as shallow as 0 . 1% . These results indicate that the proposed mechanism for neutrophil chemotaxis is able to sense very shallow gradients of chemoattractant and to amplify the response . Experimentally , neutrophils have been shown to properly orient in gradients as shallow 1% in an optimal background concentration [39] . At gradient strengths below this value , neutrophils will polarize in random directions . The likely reason for this higher threshold at 1% is that stochastic fluctuations in the chemoattractant field or pathway dynamics mask weak gradient signals or create transient gradients of similar or greater magnitude . Because our model is deterministic , these fluctuations are ignored in our simulations , so there is no physical barrier to the cell sensing shallow gradients . We also varied the background concentration , bL , from 10−4 to 102 while leaving the gradient strength fixed at gL = 10% . The results are shown in Figure 5B . At background concentrations less than 0 . 1 , the cell is unable to polarize . Conversely , at all concentrations above this threshold , the cell will polarize in the direction of the gradient . Once again , these results demonstrate that the model is able to sense gradients over a wide range of background concentrations . However , when these concentrations become saturating , we do not expect that the cell will be able to properly orient in the direction of the gradient , because molecular fluctuations at the level of receptor–ligand binding will mask any differences in the amount of chemoattractant-bound receptors across the length of the cell due to the gradient . Stochastic fluctuations are again hypothesized to create a physical barrier to gradient sensing at high background concentrations . Experimentally , chemotaxis responds biphasically to increasing concentrations of chemoattractant at fixed gradient strengths [39] . At low concentrations , insufficient chemoattractant is present to stimulate the cells . At saturating concentrations , fluctuations dominate and the cells polarize in random directions . Only at intermediate concentrations do the cells polarize correctly in the direction of the gradient . Unlike in Dictyostelium [40] , inhibition of actin polymerization prevents neutrophils from correctly polarizing in response to a gradient . In particular , RhoA is no longer restricted to the rear of the cell and the PIP3 distribution is not as sharp [18 , 22] . We also observe similar behavior in our model when the actin polymerization reaction is removed ( Figure 6A ) . In this scenario , RhoA preferentially localizes to the front of the cell as it is no longer excluded by actin polymerization . Furthermore , due to the lack of exclusion by actin polymerization , PTEN no longer localizes to the rear and is uniformly distributed along the cell membrane . The net result is that PI3K only weakly localizes to the front and , consequently , PIP3 localization is significantly attenuated ( Figure 6B , Video S2 ) . We next considered the effect of a positive feedback loop operating on PIP3 . Such a loop is known to function in neutrophils . For simplicity , we modeled this feedback loop by assuming that PIP3 enhances the catalytic activity of PI3K , though it is known that additional proteins such as Cdc42 and Rac are also involved in this feedback loop [11 , 16 , 19 , 20] . Figure 7 shows the steady-state concentrations of the Ras–PI3K complex ( black ) , PIP3 ( blue ) , and actin ( red ) in response to a 50% gradient of chemoattractant . The solid lines include the effect of the positive feedback loop while the dashed lines do not . The inclusion of the PIP3 positive feedback loop does indeed amplify the front signal but is not necessary to localize the front and rear signals in our model . Several models of eukaryotic chemotaxis include mechanisms for perfect adaptation of PIP3 levels . In other words , the PIP3 response to a uniform increase or decrease of chemoattractant is transient and eventually returns to prestimulus levels [24 , 26 , 27 , 29 , 33 , 41] . Perfect adaptation extends the range of concentrations that a cell can respond to and is generally thought to occur through the action of a global inhibitor [42] . While perfect adaptation has been confirmed in bacterial gradient sensing [43] , the experimental evidence in eukaryotes is mixed and suggests that only partial adaptation occurs [18 , 36 , 44] . In our model , the spatial and temporal response of the pathway partially adapts to a uniform increase in attractant ( Figure 8A ) . Initially , there is a rise in PIP3 because PI3K has faster dynamics than PTEN . After a short lag , PTEN levels rise , causing PIP3 levels to partially , but not completely , adapt in response to the stimuli . Figure 8B shows the dynamics of adaptation at one point on the cell's periphery . We also explored the dynamics of adaptation in response to different concentrations of chemoattractant . The results , shown in Figure S1 , demonstrate that as the concentration of chemoattractant increases , the peak response in PIP3 levels increases , as does the steady-state levels . In addition to adaptation , neutrophils can also spontaneously polarize when exposed to spatially uniform concentrations of chemoattractant [5] . We hypothesize that this polarization is due to random fluctuations in the pathway or concentration field that lead to transient asymmetries , which are subsequently amplified by the internal pathway dynamics . To explore this hypothesis , we simulated the model in response to a uniform concentration of ligand and also included a perturbation to the receptor–ligand complex in order to mimic biochemical noise . Figure 9A shows that this perturbation leads to the spontaneous polarization of both the front and back signals . Both the magnitude of the perturbation and concentration of chemoattractant affect the response . Figure 9B shows the steady-state response of the model to perturbations of varying magnitudes . The results demonstrate that the perturbation must exceed a certain threshold for the cell to spontaneously polarize . Increasing the concentration of chemoattractant decreases this threshold ( unpublished data ) . Previous work stated that a global inhibitor is necessary for a minimal model to display spontaneous polarization without lock-on behavior [30] . Our model displays this behavior without a global inhibitor , though we do not claim that our model is “minimal . ” Note that two different responses can result from uniform stimulation in our model: adaptation and spontaneous polarization . Both responses are also observed experimentally [20] . Double micropipette experiments have been proposed as a way to invalidate directional sensing models [25 , 40] . Figure 10 shows the steady-state response of our model to two gradients of varying intensity , where the abscissa denotes the percentage difference in gradient intensity . When the two gradients are of roughly the same intensity ( bL = 1 and gL = 50 ) , two fronts will form . However , when one is greater , then only a single front will form at steady state . Initially , two fronts will form , but the second front localized in the direction of the weaker gradient will eventually disappear ( unpublished data ) . The difference in gradient intensities needed for only a single front to form is a function of the gradient strength ( gL ) and background concentration ( bL ) . If either the nominal gradient strength is increased or the background concentration is decreased , then the range of intensities over which two fronts are maintained increases . We also explored the response to three gradients of equal intensity equidistantly spaced around the cell . The results are shown in Figure 11 . Three fronts of equal intensity form . However , after 50 time points , the three fronts collapse into a single front . Similar results are observed when additional gradients are added . To our knowledge , these experiments have not yet been performed using either neutrophils or HL-60s , though actin-inhibited Dictyostelium form two fronts in response to two opposing gradients of equal strength [40] . We next tested the robustness of our modeling results to changes in parameter values . While the ability of our model to correctly polarize in response to gradients was robust to most kinetic parameters ( ±50% , unpublished data ) , we found that our model was acutely sensitive to the relative concentration of PI3K . As our model lacks a global inhibitor , selective recruitment of proteins to either the front or rear of the cell necessitates a limited substrate supply . In other words , our model assumes that there is an excess of membrane binding sites relative to the concentration of cytosolic PI3K . When this assumption is violated , PI3K will bind everywhere on the membrane in response to chemoattractant , even in the presence of a strong gradient . If the supply is limited , on the other hand , then the protein will bind preferentially to the areas of the membrane where the activating signal is greatest . Figure 12A shows that when the concentration of binding sites relative to cytosolic PI3K ( γPI3K ) is less than approximately 1 . 8 , PIP3 and actin both localize around the entire cell membrane and prevent myosin from binding . Similar , though reciprocal , behavior is also observed with RhoA ( γRh ) . We contrast these results with Figure 12B . Here , we varied the association rate constant of RhoA ( aRh ) over a wide range of values , and no change in the polarization profile was observed . Neutrophils likely do not fine-tune the expression of PI3K or RhoA . The lack of robustness in the model with respect to this parameter suggests that additional regulatory mechanisms are present in the chemotaxis pathway . Our model is by no means complete . A number of proteins that regulate chemotaxis are omitted from the model , including Cdc42 , PAK , and Rac [16 , 45 , 46] . Quite possibly these proteins form additional layers of regulation that ensure robust chemotaxis . Another possibility that we cannot exclude is the presence of rapidly diffusing global inhibitors . Although our model does not require a rapidly diffusing global inhibitor , the addition of global inhibitors likely can improve the robustness of the model . However , experimental evidence for such a regulatory mechanism is currently lacking . The coincidence circuit is necessary in our model to robustly polarize the cell in response to a gradient . While Ras is known to interact with PI3K [12 , 15 , 47] , details regarding the specific mechanisms for activation are not known . To account for potential alternative mechanisms of interaction between PI3K and Ras , we explored a second model for the coincidence circuit where we assumed that membrane-bound PI3K is equally active irrespective of Ras . The role of activated Ras in this alternative model is to stabilize PI3K by preventing it from prematurely disassociating from the membrane . This Ras-mediated stabilization of PI3K leads to enhanced conversion of PIP2 to PIP3 and , as a result , forms a coincidence circuit involving Ras and PI3K . Implementing this alternative model did not require any structural changes to the original mathematical equations and instead entailed only changes to the parameter values ( see Material and Methods ) . As shown in Figure 13A , this alternative model is able to correctly polarize in response to a gradient and still is responsive to changes in the gradient direction . Furthermore , the model will also spontaneously polarize in response to a uniform increase in chemoattractant when a perturbation to the receptor–ligand complex is included ( Figure 13B ) . Unlike the original model , however , the “backness” pathway will localize at the site of the perturbation in the alternate model . In the absence of the perturbation , the response will partially adapt ( unpublished data ) . With the one exception noted above , the dynamics of the alternative model are similar to the original model . One set of experiments to test the proposed mechanism for chemotaxis is to measure the response in neutrophils or HL-60 cells with either an impaired Ras GTPase–activating protein ( GAP ) or a constitutively active Ras . Our model predicts that cells with an impaired Ras GAP will respond sluggishly to changes in the direction of the gradient ( Figure 14A ) . Likewise , our model predicts that constitutively active Ras will cause “lock-on” behavior ( Figure 14B ) . Therefore , the cell will first correctly polarize towards the micropipette but then will be unable to track the micropipette as it rotates . With the alternate model for the Ras–PI3K coincidence circuit , “lock-on” behavior is observed in both scenarios ( unpublished data ) . Potentially , these experiments could be used to discriminate between the two mechanisms , though direct biochemical interrogation of the circuit would yield more definitive results .
The ability of our model to separate front and back signals depends on the PI3K coincidence circuit and the inhibition of RhoA by actin . To test the validity of this mechanism , we suggest the rotating micropipette experiments in neutrophils or HL-60 cells with an impaired Ras GAP or constitutively active Ras . As shown in Figure 14 , our model predicts that constitutively active Ras will cause “lock-on” behavior , and cells with an impaired Ras GAP will respond more slowly to changes in the direction of the gradient . Additionally , our model may be tested with a double micropipette experiment . As shown in Figure 10 , our model predicts that the cell can form two fronts when exposed to two equal and opposite sources; however , if the sources are not equal , then the cell will prioritize one signal and form only one front . Significant deviation from this predicted behavior would suggest that our model is incorrect . Finally , our model is incomplete because it lacks many known regulators of chemotaxis in neutrophils such as Cdc42 , Rac , and PAK [11 , 16 , 19 , 20] . The model also simplified actin and myosin dynamics . Our initial goal was not to develop a comprehensive model of chemotaxis but rather a parsimonious model of neutrophil chemotaxis to explore the general mechanism proposed by Bourne and colleagues . We are currently extending the model to address these limitations .
A number of simplifying assumptions were made in formulating the model . ( 1 ) The model assumes that the rate of PI3K recruitment to the membrane is proportional to the density of chemoattractant-bound receptors . The assumed mechanism for PI3K recruitment is the following . Chemoattractant-bound receptors activate the trimeric G–protein Gi and cause the dissociation of the Gβγ heterodimer . The Gβγ heterodimer then recruits PI3K to the membrane through its interaction with the p101 regulatory subunit . In the model , we simplified this mechanism by assuming that receptors directly recruit PI3K to the membrane . ( 2 ) Both Gβγ and Ras are necessary for PI3K activation , though the mechanistic details for this regulation are lacking [15] . In the model , we assume that chemoattractant-bound receptors directly recruit PI3K to the membrane . The model assumes that PI3K is only active when it binds to activated Ras on the membrane . In the alternate model for the PI3K–Ras interaction , we assumed that membrane-bound PI3K is active irrespective of Ras . However , the interaction with Ras stabilizes PI3K on the membrane , thereby indirectly enhancing the activity of PI3K . Implementing the alternate model involved changing the values of only five parameters . The parameters and values are listed in Table 1 . ( 3 ) The model explicitly ignores a number of key regulatory proteins including Gi , G12|13 , Rac , Rock , and Cdc42 . Rather , these proteins were treated implicitly in the model . For example , we assume that ligand-bound receptor directly activates Ras rather than include the intermediate messenger Gi . ( 4 ) The model assumes that the total concentration of PIP2 and PIP3 is fixed in the cell and that their relative concentrations are determined solely by the reciprocal action of PI3K and PTEN . The role of other phosphoinositide lipids and their specific isoforms are ignored in the model along with other proteins that are known to regulate their composition , such as PI5K and SHIP [48] . ( 5 ) A number of experiments demonstrate that positive feedback loops involving PIP3 , PI3K , and actin are present in neutrophils [11 , 16 , 19 , 20] . As mechanistic details are lacking , the model assumes that PIP3 directly enhances the catalytic activity of the Ras–PI3K complex . Other mechanisms are equally likely but were not explored in this work . We note that this feedback mechanism is not necessary for polarization and instead serves to amplify front-to-back differences . ( 6 ) The mechanisms for actin polymerization and myosin attachment/contraction are simplified in the model . As detailed mechanisms for both processes are lacking , we assumed for simplicity that actin and myosin bind to the membrane at a rate proportional to the local density of PIP3 and RhoA squared , respectively . Many actin-regulating proteins such as WASP require multiple signals for activation [49] . Because our model ignores many of these signaling proteins , such as Cdc42 , we assumed that PIP3 activation of actin polymerization is cooperative . Though supporting data is lacking , we also assumed the same relation between RhoA and myosin . We note that cooperativity is not necessary for the model to reproduce the results described . The model also assumes that the two events are mutually exclusive , with both proteins competing for the same binding sites on the membrane . ( 7 ) The model assumes that actin inhibits the binding of PTEN . While there is no direct evidence to support this mechanism , PTEN is known to localize with RhoA at the rear of the migrating cell [21] . Furthermore , Cdc42 appears to control the localization of PTEN as PTEN is excluded from regions where Cdc42 is active . As we do not include Cdc42 in our model , we have used actin as a proxy localization signal . ( 8 ) In our model , PI3K is necessary for actin polymerization . While inhibition of PI3K severely retards chemotaxis , it does not completely abolish it [45] . These results suggest that a parallel , PI3K-independent , pathway is also involved in regulating chemotaxis and motility [50 , 51] . As the details for these parallel pathways are lacking , we were unable to include them in the model . The model assumes for simplicity that the cell is a cylindrical disk in two dimensions . Because diffusion of cytosolic proteins is significantly greater on average than membrane-bound proteins , ( 10 μm2/s versus 0 . 05 − 0 . 5 μm2/s [52] ) , we assume that the interior of the cell is homogeneous and well-mixed . As a result , we are able to recast the model as a coupled set of one-dimensional partial differential equations with periodic boundary equations , where the spatial coordinate θ specifies the axial position on the membrane . The governing set of coupled partial differential equations for the model are the following . The definitions for the state and domain variables are given in Table 2 . As detailed measurements for the total concentrations of the different molecular species in the model are currently lacking , we chose to normalize the concentrations in the model such that they can only take values between zero and one ( e . g . , XC ( θ , t ) ∈[0 , 1] ) . Furthermore , to simplify analysis , we also reformulated the model in terms of dimensionless parameters using the scaling factors defined in Table 3 . The following transformations were used to convert the model into dimensionless form . Definitions for the dimensionless parameters along with their nominal values used in the simulations are given in Table 1 . The governing equations for the model in dimensionless form are the following . The spatial derivative operator was discretized using finite differences on a grid with 100 points , and the resulting set of coupled ordinary differential equations were solved in Matlab 7 . 2 ( The Mathworks , http://www . mathworks . com ) using the ode15s routine . Numerical solution required approximately one minute of CPU time on an AMD Athlon 2 . 2-MHz desktop computer running the Linux operating system . Matlab m-files for all figures are available at http://openwetware . org/wiki/Rao_Lab:Code . | Neutrophils target sites of infection and inflammation by sensing chemical signals produced by damaged tissue and infecting microbes and then move in the direction where their concentration is greatest . An open question is how neutrophils integrate this information to determine the direction of motility . We present a mathematical model for the intracellular signaling network regulating polarization and chemotaxis in neutrophils . We demonstrate how the activation of two antagonizing pathways robustly establishes the front and back of the migrating cell . The model is able to reproduce a number of experimental studies , and new experiments are proposed to test different aspects of the model . A key result is the characterization of a coincidence circuit involving phosphatidylinositol-3-kinase ( PI3K ) and Ras . We demonstrate that this circuit plays a critical role in selectively localizing F-actin to the front of the cell and actomyosin complexes to the rear . As directed motility in response to chemical cues is critical in a number of processes including wound healing and tumor metastasis , the results and insights gained from the model may be applicable to other cell types and organisms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology",
"homo",
"(human)",
"computational",
"biology"
] | 2007 | A Mathematical Model for Neutrophil Gradient Sensing and Polarization |
In a dynamic world , an accurate model of the environment is vital for survival , and agents ought regularly to seek out new information with which to update their world models . This aspect of behaviour is not captured well by classical theories of decision making , and the cognitive mechanisms of information seeking are poorly understood . In particular , it is not known whether information is valued only for its instrumental use , or whether humans also assign it a non-instrumental intrinsic value . To address this question , the present study assessed preference for non-instrumental information among 80 healthy participants in two experiments . Participants performed a novel information preference task in which they could choose to pay a monetary cost to receive advance information about the outcome of a monetary lottery . Importantly , acquiring information did not alter lottery outcome probabilities . We found that participants were willing to incur considerable monetary costs to acquire payoff-irrelevant information about the lottery outcome . This behaviour was well explained by a computational cognitive model in which information preference resulted from aversion to temporally prolonged uncertainty . These results strongly suggest that humans assign an intrinsic value to information in a manner inconsistent with normative accounts of decision making under uncertainty . This intrinsic value may be associated with adaptive behaviour in real-world environments by producing a bias towards exploratory and information-seeking behaviour .
In many decision situations , agents possess only incomplete information about decision outcomes , and may choose to seek out further information before choosing a course of action [1 , 2] . For instance , a surgeon considering whether to operate on a tumor might first request a biopsy to determine whether the tumor is malignant or benign . Despite being a key feature of choice problems in natural settings , information seeking is not considered within many standard accounts of decision making under risk and uncertainty [3–5] . Moreover , it has been shown that some animals choose to seek information even when that information cannot be used to improve future outcomes [6–8] . This behaviour , which is suboptimal from the perspective of expected reward maximization , suggests that biological agents may attach a value to information which is not solely defined in terms of tangible future outcomes [9] . Historically , many theories of information valuation have adopted an instrumental framework , in which the value of information is calculated solely in terms of expected instrumental benefit [10–12] . These theories predict that a decision-maker should seek information only if the information is expected to impart a tangible benefit in excess of its cost [11] . For instance , a clairvoyant charging $100 to reveal whether stock prices will rise or fall should only be consulted if a payoff greater than $100 is expected to result from using this information . Instrumental valuation of information is normatively optimal , in the sense that it maximises expected monetary reward . However , one strong prediction of instrumental valuation is that information of no instrumental use for acquiring payoffs ( henceforth termed non-instrumental information ) should not affect choice behaviour . As a result , instrumental valuation of information cannot easily explain curiosity-driven or purely exploratory behaviours [13 , 14] . An alternative proposal is that biological agents may attach an intrinsic value to information , such that information about relevant future outcomes is valued for its own sake , independent of direct , tangible payoffs [15] . Similarly , economic decision theory has posited that humans might possess a preference for early resolution of uncertainty which would result in intrinsic value of information [16–18] , and recent theories of active inference propose that choice behaviour can be explained by sensitivity to information gain as well as to extrinsic reward [19] . In support of intrinsic valuation of information , human participants have been shown to prefer early to late information about receiving an unavoidable electric shock [20] , and to be conditioned by non-instrumental information in a behavioural conditioning paradigm [21] . Moreover , neural data from humans and non-human primates have shown that non-instrumental information is encoded using similar mechanisms , and within similar circuits , to primary and monetary reward [9 , 22–24] . These findings are consistent with the hypothesis that biological agents assign an intrinsic reward value to non-instrumental information about future outcomes using a coding scheme commensurate with primary and monetary reward . One limitation of previous empirical work assessing preference for information in humans is that information available in decision-making tasks is usually of instrumental benefit to participants , such that it is difficult to dissociate the intrinsic value of information from its instrumental value [25 , 26] . To address this issue , the present study adapted a task from the animal literature which allowed preferences for non-instrumental information to be elicited in a well-controlled and incentive-compatible manner [22] . Using this task , we sought to test one counterintuitive prediction of intrinsic valuation of information: that , like starlings and pigeons , human participants would trade off information against extrinsic reward by sacrificing part of an uncertain future reward in exchange for early but non-instrumental information about reward likelihood [27] . Furthermore , among theories positing an intrinsic value of information , the source of this value is often unspecified . For instance , the Kreps-Porteus model in economic decision theory predicts a preference for early resolution of uncertainty from a particular axiomatic formulation of utility , but does not specify a cognitive mechanism which might drive this preference [16] . One proposal is that preference for non-instrumental information might result from an aversion to temporally prolonged uncertainty , such that agents may seek information in order to obtain relief from uncertainty [13 , 28–30] . We therefore tested a novel computational cognitive model , which assumed that inter-individual variability in the intrinsic value of information resulted from stable trait-level individual differences in degree of aversion to uncertainty , against a standard expected reward maximization model , which assumed that information was assigned solely instrumental value [11] . Finally , in order to determine whether the duration of uncertainty affected participants’ preference for information , we also conducted an additional experiment in the rate at which non-instrumental information was delivered was experimentally manipulated .
Experiment 1 assessed participants’ willingness to forfeit monetary reward in exchange for non-instrumental information , and examined the effect of information cost on information preference . Across cost conditions , participants chose the informative stimulus on 43 . 95 percent of trials ( SD = 20 . 28 ) , while showing good task engagement as evidenced by a low proportion of missed responses ( M = 1 . 67 percent , SD = 1 . 72 ) . On average , across cost conditions participants sacrificed 2 . 87 percent of available winnings in exchange for early information about the lottery outcome ( SD = 3 . 21 ) . A one-way repeated-measures analysis of variance ( ANOVA ) revealed that choice proportions were modulated by the cost of information ( F ( 1 . 89 , 73 . 60 ) = 65 . 68 , p < . 001; partial η2 = 0 . 63; see Fig 2a ) , with information choice proportion monotonically decreasing with increases in information cost . Control analyses revealed that behaviour was not significantly affected by the key used to select responses ( left versus right arrow: t ( 39 ) = -0 . 71 , p = . 24 ) or the nominal identity of the informative stimulus ( A versus B: t ( 39 ) = 1 . 40 , p = . 08 ) . We next used post-hoc t-tests with Bonferroni correction to assess whether participants’ behavior was consistent with expected reward maximization . Expected reward maximization , which implies solely instrumental valuation of information , predicts that participants should be indifferent between the informative and non-informative stimulus when information is free , and that the non-informative stimulus should dominate the informative stimulus for any non-zero information cost [11] . In the zero-cost condition , informative stimulus choice proportion was significantly greater than the indifference point of 0 . 5 ( t ( 39 ) = 16 . 83 , p < . 001 ) . In each of the non-zero cost conditions , informative stimulus choice proportion was significantly greater than zero ( 1-cent condition: t ( 39 ) = 7 . 17 , p < . 001; 3-cent: t ( 39 ) = 5 . 41 , p < . 001; 5-cent: t ( 39 ) = 4 . 76 , p < . 001 ) . These results indicate that participants sacrificed future reward for early information , which is inconsistent with expected reward maximization . In addition , we observed notable individual differences in patterns of information seeking behaviour ( see Fig 2b ) , indicating heterogeneity of task strategies between participants . To formalise the comparison between instrumental and intrinsic theories of information valuation , we implemented these theories as competing computational cognitive models , and assessed which model provided the best account of both group- and individual-level data . The two models we assessed were termed the Expected Value of Information ( EVI ) model , which incorporated solely instrumental valuations of information , and the Uncertainty Penalty ( UP ) model , which also incorporated intrinsic valuation of information by assuming that the source of information’s intrinsic value was an aversion to temporally prolonged uncertainty [13 , 28–30] . Both models considered the task in a Markov Decision Process ( MDP ) framework , and differed only in choice of state value function [31] ( see also Experimental Protocols ) . We found that , in addition to providing the best overall account of choices across participants ( smallest overall Bayesian Information Criterion ( BIC ) value; see Table 1 and Fig 3 ) , the UP model provided the best fit for a large majority of individual participants . Accordingly , a likelihood-ratio test revealed that including the participant-specific uncertainty penalty parameter k greatly improved the overall fit of the UP model relative to the EVI model ( χ2 ( 40 ) = 1338 . 34 , p < . 001 ) . Moreover , the UP model provided an unbiased fit to the data of all participants , including those who displayed a relatively weak overall preference for information ( see Fig 4 ) . By contrast , the EVI model systematically underestimated informative stimulus choice proportions across all participants . Furthermore , we found that the best-fitting values of the UP’s scaling parameter k were greater than zero across participants ( Wilcoxon signed-rank test: Z = 5 . 51 , p < . 001 ) and , in addition , were strongly correlated with overall proportion of information-seeking choices across participants ( Spearman’s rho = 0 . 95 , p < . 001 ) . This indicates that participants with a stronger aversion to uncertainty ( higher k values ) assigned a greater intrinsic value to information , and therefore made more information-seeking choices ( see also SI section 1 for individual parameter estimates and parameter-behaviour correlations ) . Although unsurprising given the structural design of the UP model , the strength of this relationship serves to demonstrate that the UP model parameter designed to capture individual differences in information preference succeeded in doing so . In addition , since the UP model’s implementation of intrinsic valuation of information assumes that aversion to uncertainty is a stable trait of participants , a secondary prediction of this model is that information’s intrinsic value ought to be stable across time for each participant . In order to test this prediction , we calculated information choice proportion separately in each of the seven experimental blocks in Experiment 1 , and assessed the effect of block number on information preference using a 4×7 repeated-measures ANOVA with within-subjects factors of information cost ( 0 , 1 , 3 , 5 cents ) and task block ( 1 to 7 ) . We found no significant main effect of task block on information choice proportion ( F ( 3 , 117 . 37 ) = 1 . 73 , p = . 16 ) and no significant interaction between task block and information cost ( F ( 9 . 58 , 373 . 59 ) = 1 . 05 , p = . 40 ) . These results indicate that informative stimulus choice proportions did not differ significantly across the task ( see Fig 5 ) , as predicted by the UP model . Finally , we performed an additional control analysis to ensure that the relative advantage of the UP model relative to the EVI model was not simply due to its better performance in the zero-cost condition . To this end , we repeated the model-fitting procedure while excluding all trials in the zero-cost condition ( that is , the models were fit solely on the basis of the 1 , 3 , and 5-cent cost conditions ) . Once again , we found that the UP model ( BIC = 718 . 31 ) provided a significantly better fit to data than the EVI model ( BIC = 1031 . 60; likelihood ratio test: χ2 ( 40 ) = 489 . 88 , p < . 001 ) . This indicates that the UP model , which incorporated intrinsic valuation of information , outperformed the EVI model even when zero-cost trials were excluded from analysis . A logical consequence of intrinsic valuation of information is that the temporal profile of uncertainty ought to affect participants’ preference for observing an informative stimulus . In particular , the same amount of information should have a different value to participants depending on the rate at which it resolves uncertainty . In Experiment 2 , we tested this prediction among a new sample of participants . In Experiment 2 , cards could be revealed at a rate of either 1 , 3 or 5 seconds per card in each trial , instead of a constant rate of 3 seconds per card as in Experiment 1 . Participants in Experiment 2 made information-seeking choices on 42 . 89 percent of trials ( SD = 24 . 72 ) , indicating comparable task performance to Experiment 1 . Participants showed good levels of task engagement , with low levels of missed responses ( M = 1 . 11 percent , SD = 1 . 87 ) . On average , participants sacrificed 2 . 75 percent of available winnings in exchange for early information about the lottery outcome ( SD = 3 . 49 ) . A 4×3 repeated-measures ANOVA was used to assess the effect of information cost ( 0 , 1 , 3 , 5 cents ) and information rate ( 1 , 3 , 5 seconds per card ) on information-seeking choice proportions . The significant main effect of information cost on information-seeking choice proportions was replicated , ( F ( 1 . 86 , 72 . 70 ) = 84 . 16 , p < . 001; partial η2 = 0 . 68 ) , and , crucially , we also found a significant main effect of information rate on information-seeking choice proportions ( F ( 2 , 78 ) = 3 . 60 , p < . 05; partial η2 = 0 . 08; see Fig 6 ) . This indicates that behaviour was modulated by the rate as well as the cost of information , though the effect size of information rate was substantially smaller than the effect size of information cost . There was no significant interaction between information cost and information rate ( F ( 6 , 234 ) = 1 . 85 , p = . 09 ) although there was a non-significant trend for the effect of information rate to be larger in the positive cost conditions ( 1 , 3 , and 5 cents ) than in the zero-cost condition . One potential explanation of this effect is that participants may have preferred sooner rather than later resolution of uncertainty . To explore this possibility , we formulated and compared additional computational models assuming temporal discounting of future states . For most participants , these models did not provide a better fit to data than the undiscounted UP and EVI models . However , the degree to which participants discounted future information was associated with individual differences in the effect size of information rate ( see S3 Text ) .
Participants in the present study consistently preferred an informative stimulus to a perceptually equivalent non-informative stimulus , despite the fact that information could not be used to improve future outcomes . Moreover , in many cases participants were willing to sacrifice future monetary reward in exchange for this early but non-instrumental information . Since the non-informative stimulus was always of equal or greater expected monetary reward , this pattern of results strongly suggests that participants assigned an intrinsic value to information . This stands in contrast to predictions of instrumental theories of information valuation based on expected reward maximization [10–12] , but is consistent with the preference for early resolution of uncertainty posited by decision theory [16–18] , and with the behavioural sensitivity to information gain proposed by active inference [19] . Although it has previously been conjectured that intrinsic valuation of information may result in willingness to pay for payoff-irrelevant information [27] , this effect has not previously been demonstrated in humans using a well-controlled cognitive task . We found that the UP model , a novel computational model of information seeking , provided a good account of intrinsic valuation of information by assuming that preference for information resulted from aversion to temporally prolonged uncertainty [13 , 28–30] . Notably , we found that the model was able to capture individual differences in strength of information preference across participants , as well as providing a good account of group-level results . It is important to note that aversion to temporally prolonged uncertainty as implemented in the UP model is mathematically and conceptually distinct from the economic concept of risk aversion [32] . Risk aversion as commonly understood cannot predict the preference for information exhibited by participants in the present task , since at the point of choice informative and non-informative stimuli were associated with identical outcome probabilities , and differed only in the rate at which outcome uncertainty was resolved . Similarly , although the informative stimulus was associated with reduced payoff variance in non-zero cost conditions , a simple mean-variance tradeoff [33] does not provide a coherent account of preference for information either , since participants’ information preference was strongest in the zero-cost condition , where both mean and variance of payoffs were identical for the two stimuli . Notwithstanding this result , however , we also found that the UP model fit data well even when trials in the zero-cost condition were excluded from analysis , thus giving us confidence that participants assigned an intrinsic value to information in both zero and non-zero information cost conditions . Furthermore , consistent with the theory that information valuation is a stable trait-level feature of individuals , we found that information preference was stable across time within the task . This would not have been expected if , for instance , participants only sought information in order to learn payoff contingencies in early blocks of the task . In addition , we found that preference for information was modulated by the rate at which uncertainty was resolved , such that participants exhibited a stronger preference for non-instrumental information when information was delivered at a faster rate . This result is analogous to the preference for faster monetary reward rate in choice behaviour [34] . Moreover , although the effect of information rate cannot be directly captured within the UP model , the direction of the information rate effect is consistent with discounting of future information , analogous to the temporal discounting of future rewards in human judgment and decision making [35] . As such , the results of the present study are in line with the proposal that humans treat information as though it has an intrinsic reward value commensurable with ( and perhaps encoded in the same neural circuits as ) primary and monetary reward [2 , 15 , 22] . However , we also note that the results of Experiment 2 demonstrated a relatively small effect size of information rate; future research should therefore further investigate the nature and robustness of this effect . Participants in the present study sacrificed future monetary reward in exchange for early but payoff-irrelevant information . This behaviour , which is suboptimal from the perspective of expected reward maximization , has previously been observed in pigeons and starlings [6–8] . In the present study we present for the first time a well-controlled cognitive paradigm with which to assess this behaviour in humans . We note that previous studies in human participants have reported results generally consistent with a willingness to pay for early resolution of uncertainty , such as a greater preference for a risky lottery whose uncertainty was resolved immediately relative to an equivalent lottery whose uncertainty was resolved gradually [36 , 37] , and a willingness to pay for immediate resolution of uncertainty rather than a 50 percent probability of delayed resolution of uncertainty [38] . Among cognitive studies explicitly assessing the value of non-instrumental information , Pierson & Goodman ( 2014 ) found that participants self-reported a willingness to pay for non-instrumental information [39] . However , this behaviour was only assessed using a hypothetical survey task , which may have confounded results given the well-documented disparity in behaviour between hypothetical and incentive-compatible choice tasks [40] . Separately , a behavioural economic study using an incentive-compatible task concluded that observing non-instrumental information was related not to intrinsic valuation of information per se , but to a desire to increase one’s post-hoc confidence regarding an earlier decision [41] . This explanation predicts that participants will only seek non-instrumental information if it provides feedback on an earlier decision . Our results are inconsistent with this explanation , since no such decision was present in the task used in the present study . The strength of our conclusions is based on a well-controlled task in which informative and non-informative stimuli were perceptually identical , and in which preferences for information were elicited in a fully incentive-compatible fashion . The results of the present study are also conceptually consistent with the preference for early resolution of uncertainty described in economic decision theory by the Kreps-Porteus model [16] , and used to account for anomalous patterns of stock pricing in finance by Epstein and Zin [18] . Our empirical and computational findings complement these theories: whereas the Kreps-Porteus model demonstrates that preference for early resolution of uncertainty is a consequence of a particular formulation of recursive utility , in the present study we present a cognitive process model which provides a psychologically plausible account of information-seeking behaviour . Specifically , our results provide evidence that information seeking may result from an aversion to temporally prolonged uncertainty [13 , 28–30] . One interesting finding in this respect was that there was a negative correlation across participants between the UP model’s information preference parameter k and its response stochasticity parameter β . This correlation was such that participants who assigned a stronger intrinsic value to information also tended to exhibit greater response stochasticity . This relationship is of theoretical interest , since it has been proposed that information-seeking behaviour may result from high levels of response stochasticity in exploration-exploitation dilemmas ( e . g . [25] , but see also [42] ) , or via ε-greedy action selection methods in reinforcement learning [43] . Although the superior goodness-of-fit of the UP model in the present study clearly indicates that response stochasticity alone cannot account for participants’ information-seeking choices , the correlation between k and β raises the interesting possibility that intrinsic valuation of information and response stochasticity may make separable but related contributions to exploratory behaviours . Under this hypothesis , the k parameter would correspond to directed exploration , a goal-directed process aimed specifically at reducing uncertainty , whereas the β parameter would correspond to a more diffuse form of undirected exploration . Future research should further investigate this hypothesis . However , it is also important to note that behaviour could also be explained by an appetitive drive for information as well as an aversion to uncertainty . Because , according to information theory , uncertainty and information are mathematical conjugates [44] , aversion to uncertainty makes similar behavioural predictions to an appetitive desire for information . Accordingly , it is possible to reparametrise the UP model to explain behaviour in terms of an information value bonus , rather than an uncertainty penalty , with equivalent behavioural predictions . As such , behavioural data alone may not be sufficient to distinguish between behaviour driven by uncertainty aversion and behaviour driven by an appetitive desire for information . One possibility for future research is that , since appetitive and aversive stimuli are processed in distinct neural circuits [45] , it may be possible to use neural recordings to disentangle these two potential cognitive mechanisms for information valuation . In accounting for the results of the present study we have primarily drawn upon theories proposing that intrinsic valuation of information can result from an aversion to temporally prolonged uncertainty [18 , 28–30] . However , alternative theoretical frameworks can also account for the present study’s results in terms of a positive information bonus ( consistent with an alternate parametrisation of the UP model described below ) . For instance , it has been proposed that agents may derive utility from maintaining an internal model of the environment which is well-adapted to the statistics of natural stimuli [46 , 47] . A natural consequence of this model is that agents should place a non-zero value on information about the external environment , even when no behaviour can be directly conditioned on this information ( for instance , an intrinsic curiosity reward , as proposed by Schmidhuber , 2009 [48] ) . Similar intuitions regarding the appetitive value of information have been formalised in several general theories of cognition , including active inference theory [19 , 47] and optimal Bayesian exploration [49] . Such theories can be extended to account for seemingly paradoxical attitudes towards information in other settings , such as participants’ preference for maximising entropy over choice options as well as simply maximising expected reward [50] , as well as seemingly paradoxical patterns of self-deception in financial choices [51] . At a neurocomputational level , appetitive valuation of information is also consistent with the notion of dopaminergic novelty or exploration bonuses [52] . Notwithstanding the above , however , a further possibility proposed by Beierholm and Dayan ( 2010 ) is that an apparent preference for informative stimuli might , in fact , be driven by task disengagement , leading to a relatively greater decrease in the subjective value of the non-informative stimulus [53] . The paradigm tested in the present study sought to prevent such task disengagement by means of pseudo-randomly occurring ‘catch trials’ , in which participants were required to make a rapid button-press response to one of the cards in either the informative or the non-informative stimulus . This manipulation helped to ensure that participants maintained task engagement even when observing the non-informative stimulus . Although this cannot conclusively rule out the possibility that participants were somewhat more engaged by the informative than the non-informative stimulus , it does ensure that participants could not fully disengage from the task during observation of non-informative stimuli . Moreover , the behavioural paradigm that we tested allows for well-controlled manipulation of task engagement: by increasing or decreasing the frequency of catch trials , it should be possible to manipulate the degree to which participants disengage during viewing of the non-informative stimulus . The Beierholm and Dayan model makes the empirical prediction , which could be tested in future research , that information preference ought to be strongest for greater degrees of task disengagement ( that is , low catch trial frequency ) , and that information preference ought to decrease in strength with increasing catch trial frequency . These theoretical caveats notwithstanding , the results of the present study provide clear behavioural evidence that human participants derive utility from non-instrumental information in a manner inconsistent with traditional models of information valuation . The behavioural task assessed in the present study provides a well-controlled means for assessing intrinsic valuation of non-instrumental information , and the UP model allows for individual differences in the strength of information valuation to be quantified in a principled and mathematically tractable fashion . Since intolerance of uncertainty has been proposed as a trans-diagnostic treatment marker for emotional disorders [54] , understanding information-seeking behaviours may shed light on the symptomatology of disorders including generalised anxiety disorder and obsessive compulsive disorder [55] . For instance , the compulsive checking behaviours which are a hallmark of obsessive compulsive disorder may represent a form of pathological information-seeking behaviour . From this perspective , it might be possible to redescribe some behavioural features of obsessive compulsive disorder as an excessive intrinsic valuation of information driven by excessive levels of aversion to uncertainty . As such , we would hypothesise that individuals with obsessive compulsive disorder would exhibit a high willingness to pay for non-instrumental information in the task used in the present study . The results of the present study also have bearing on studies of the exploration-exploitation dilemma , in which participants trade off information seeking and reward seeking [25] . A common finding in this literature is that participants seek out more information than is optimal [26] . Our results may help shed light on this finding: intrinsic valuation of information may cause participants to place a premium on information , resulting in a valuation of information in excess of its purely instrumental value . More broadly , we note that although preference for information in the present task was suboptimal from the restricted perspective of monetary reward maximisation , intrinsic valuation of information may be adaptive in more naturalistic environments . Choices in natural settings often resemble dynamic constrained optimisation problems , in that organisms are presented with epistemic uncertainty and poorly defined action-outcome contingencies . In these environments , the instrumental value of seeking information may be computationally intractable , and intrinsic valuation of information might induce a bias toward gathering information that encourages exploratory behaviour even when the usefulness of that exploratory behaviour is not immediately clear . As such , intrinsic valuation of information may induce patterns of behaviour akin to an exploration or novelty bonus [52] . Therefore , in dynamic and uncertain environments intrinsic valuation of information may be associated with profound long-run benefit , in spite of locally suboptimal outcomes in artificial task environments such as that employed by the present study . More broadly , we do not propose that there exists any single level of intrinsic information valuation that will produce optimal behaviour across all environmental conditions . For instance , a strong intrinsic valuation of information may be beneficial when exploration costs are low and overall uncertainty is high , but result in suboptimal performance in situations where exploration is relatively expensive , or where overall environmental uncertainty is low . In summary , our results provide strong evidence for intrinsic valuation of information in humans , and we present a novel cognitive process model which suggests that aversion to prolonged uncertainty may be an important psychological determinant of this value . We show that intrinsic valuation of information can result in seemingly suboptimal behaviours , such as a willingness to sacrifice future monetary reward in exchange for immediate but unusable information about relevant future outcomes . More broadly , our results provide a plausible psychological mechanism for human curiosity and exploration , and may explain features of decision making under uncertainty that have hitherto been considered irrational .
Participants were staff and students of the University of Melbourne . In Experiment 1 , we recruited forty-one participants ( 15 male , 26 female; 40 right-handed , 1 left-handed ) , aged 18 to 31 ( M = 22 . 28 , SD = 2 . 63 ) . In Experiment 2 , we recruited 40 participants ( 14 male , 26 female; all right-handed ) aged 18 to 32 ( M = 22 . 90 , SD = 4 . 04 ) . Participants gave voluntary informed consent , research was conducted in accordance with the Declaration of Helsinki , and protocols were approved by the University of Melbourne Human Research Ethics Committee ( ID 1341084 ) . As compensation for participation , participants received a flat payment of AUD $10 plus all lottery winnings ( lottery winnings in Experiment 1: M = $9 . 10 , SD = $1 . 86; Experiment 2: M = $7 . 15 , SD = $0 . 91 ) . Stimuli were presented using the Psychophysics Toolbox [56] and MATLAB R2012b ( The Mathworks , Natick , MA ) on a Macintosh Mini connected to an LCD monitor with resolution 1920×1080 pixels at a screen refresh rate of 60Hz . Experiment 1 comprised seven blocks , each consisting of sixteen trials total: four trials in each of the four cost conditions ( 0 , 1 , 3 , 5 cents ) , with win probabilities pseudo-randomised to ensure that win rates for each cost condition were identical . As is standard practice in computational modelling studies , participants were randomly assigned to one of four pre-generated trial sequences . Participants completed the task in approximately 1 hour . In Experiment 2 , the rate at which cards were revealed differed between blocks . In each block , information could be revealed at a rate of either 1 , 3 , or 5 seconds per card . As such , the lottery delay period varied across blocks ( 6 seconds total in 1 sec/card blocks , 18 seconds for 3 sec/card blocks , 30 seconds for 5 sec/card blocks ) . Participants completed 6 blocks of 12 trials each . Each participant was assigned to one of three counterbalanced trial orders , in which no two adjacent blocks belonged to the same information rate condition . Participants completed the task in approximately forty minutes . The primary dependent variable Pr ( Info ) was the proportion of all choices ( excluding missed responses ) in which participants elected to observe the informative stimulus . To ensure that participants maintained task engagement and attended to each stimulus type equally , approximately 10 per cent of all trials were designated as catch trials . In catch trials , instead of revealing a black or red card , one of the cards was revealed to be a white X , to which participants responded by pressing any key within 1 . 5 seconds . A successful response led to progression to the subsequent trial without penalty; failure to respond resulted in a $1 penalty . Participants who failed to respond to more than two catch trials across the experiment were excluded from all further analyses . This resulted in the exclusion of one participant in Experiment 1 ( successful catch trial responses in Experiment 1: M = 96 . 88% , SD = 5 . 56%; Experiment 2: M = 97 . 5% , SD = 5 . 80% ) . Rates of successful responses to catch trials did not differ significantly between informative and non-informative stimuli ( Experiment 1: t ( 37 ) = 0 . 74 , p = . 46; Experiment 2: t ( 37 ) = 0 . 61 . p = . 55 ) . There is therefore no evidence to suggest that participants’ catch trial performance differed as a function of stimulus type . Models represented the task as an MDP , in which each trial was a decision problem with two actions ( informative/non-informative stimuli ) , discrete states corresponding to different configurations of red and black cards ( see Fig 7 ) , and state transition probabilities corresponding to relative probabilities of red/black cards . Using dynamic programming , we calculated the action value of observing each of the two stimuli under varying assumptions about the nature of valuation , and used these action values to predict choice proportions . Competing models used identical MDP task representations , and differed only in the definition of the equation used to calculate action values . The information seeking task used in the present study can be formally characterised as follows: in each trial x , participants chose an action ax from the set A = {I , N} , where I denotes a choice to observe the informative signal and N denotes a choice to observe the non-informative signal . The outcome of each trial was denoted yx and could be either a win , yx = 1 , with probability Pblack , or a loss , yx = 0 , with probability Pred . By definition , Pred = 1—Pblack , and in the present study , it was always the case that Pblack = Pred = 0 . 5 . Let cx denote the cost in cents of observing the informative signal on a given trial , drawn from the set C = {0 , 1 , 3 , 5} . Each trial’s winnings , denoted rx , depended only on the action selected and the predetermined outcomes of the trial lottery , such that rx={20−cx , ax=I and yx=120 , ax=N and yx=10 , otherwise ( 1 ) The task’s structure was implemented as an MDP by considering every different possible configuration of red and black cards as a separate state . This is a natural way of discretising trials of the task such that each state represents a perceptually distinct epoch within the trial . The general structure can be described as follows: depending on the action selected , a participant traverses one of two state trees , corresponding to the two signal types . As such , the two state trees are structurally identical and differ only in the sense that the final card state reached in the informative signal tree perfectly predicts whether the lottery outcome will be a win or a loss , whereas the final card state in the non-informative signal tree may transition to either outcome . Since transitions within the state trees depend on the relative likelihood of drawing red and black cards , state transitions are governed by the probabilities Pblack and Pred . This structure is illustrated schematically in Fig 7 . The structure and parameters described above give a complete description of the states , actions , rewards , and state transition probabilities of the information seeking task . As a result , standard analytic techniques of MDPs can be applied to solve this decision problem . Specifically , using dynamic programming [57] , it is possible to calculate the action value Q of each of the actions I and N , and to use these action values to predict choice proportions for the two actions . Each of the competing behavioural models used an identical MDP framework , and models differed only in how the action values are calculated . More precisely , in each of the models assessed , action values were calculated by solving a Bellman optimality equation , where only the definition of this equation and its free parameters varied between models . Model fitting procedures were identical for each of the models . | Acquiring information about the external world is vital for planning and decision making . However , recent research has shown that some animals choose to acquire information at considerable cost even when the information is of no practical benefit , a counter-intuitive behavior associated with suboptimal outcomes . In this study , we demonstrate that humans also engage in this suboptimal behavior by forfeiting future monetary reward in exchange for early but unusable information about future outcomes . Our results suggest that participants attach a value to information beyond its purely instrumental value . This preference for information may help account for apparent anomalies in human choice behavior such as compulsive checking behaviors in obsessive-compulsive disorder , and excessive and wasteful use of uninformative laboratory testing in hospitals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"decision",
"theory",
"medicine",
"and",
"health",
"sciences",
"decision",
"making",
"applied",
"mathematics",
"social",
"sciences",
"neuroscience",
"simulation",
"and",
"modeling",
"systems",
"science",
"mathematics",
"statistics",
"(mathematics)",
"cognition",
"behavior... | 2016 | Intrinsic Valuation of Information in Decision Making under Uncertainty |
Multiscale modeling provides a very powerful means of studying complex biological systems . An important component of this strategy involves coarse-grained ( CG ) simplifications of regions of the system , which allow effective exploration of complex systems . Here we studied aspects of CG modeling of the human zinc transporter ZnT2 . Zinc is an essential trace element with 10% of the proteins in the human proteome capable of zinc binding . Thus , zinc deficiency or impairment of zinc homeostasis disrupt key cellular functions . Mammalian zinc transport proceeds via two transporter families: ZnT and ZIP; however , little is known about the zinc permeation pathway through these transporters . As a step towards this end , we herein undertook comprehensive computational analyses employing multiscale techniques , focusing on the human zinc transporter ZnT2 and its bacterial homologue , YiiP . Energy calculations revealed a favorable pathway for zinc translocation via alternating access . We then identified key residues presumably involved in the passage of zinc ions through ZnT2 and YiiP , and functionally validated their role in zinc transport using site-directed mutagenesis of ZnT2 residues . Finally , we use a CG Monte Carlo simulation approach to sample the transition between the inward-facing and the outward-facing states . We present our structural models of the inward- and outward-facing conformations of ZnT2 as a blueprint prototype of the transporter conformations , including the putative permeation pathway and participating residues . The insights gained from this study may facilitate the delineation of the pathways of other zinc transporters , laying the foundations for the molecular basis underlying ion permeation . This may possibly facilitate the development of therapeutic interventions in pathological states associated with zinc deficiency and other disorders based on loss-of-function mutations in solute carriers .
Multiscale computer simulations provide a general philosophy that allows one to explore complex systems while choosing the proper level of details for different regions of the system [1] . Thus , such approaches use coarse-grained ( CG ) treatments where molecular systems are simplified by , for example , treating groups of atoms as a single particle , to decrease the required resources and allow longer or larger simulations . The success of such systems in predicting and explaining biological phenomena has been exemplified by many studies of complex systems ( e . g . [1–5] ) . Studying a system at CG-level resolution is particularly beneficial when the systems are very large , the process investigated takes place over long time-scales ( ≥ microseconds ) , or when the system is of low resolution . In the latter case , the advantage of CG modeling is that it does not treat all atoms explicitly , and therefore low-resolution structures ( commonly cryo-EM ) or models ( with the accompanied uncertainties and inaccuracies ) are good examples where CG modeling can perform better than full atomistic simulations ( e . g . [6] ) . In fact , in studies of complex systems it is recommended to start by charting the system with CG modeling . Lastly , the advantage of CG is that it results in a smoother landscape and thus results in faster convergence , which is arguably one of the key requirements in computational biology simulations . In the current study we demonstrate the use of several aspects of multiscale modeling by studying the physiologically important human zinc transporter ZnT2 and investigating its permeation pathway . Zinc is the second most abundant trace element in the human body and it is estimated that over 10% of the proteins in the human proteome are capable of zinc binding [7] . Zinc is crucially important for numerous physiological processes including metabolism of nucleic acids , regulation of gene expression , signal transduction , cell division , immune- and nervous-system function , wound healing and apoptosis [8] . In humans , intracellular zinc homeostasis is tightly regulated via the transport functions of two transporter families containing 24 different transmembrane carriers , ZIP1-14 and ZnT1-10 [9] . Moreover , various metalloproteins bind free zinc , hence buffering cytoplasmic zinc levels [10] . In the past decade , the human zinc transporter 2 ( ZnT2/SLC30A2 ) was found to be the predominant transporter mediating the translocation of zinc into breast milk in lactating mammary epithelial cells [11] , involved in clinical cases of transient neonatal zinc deficiency ( TNZD ) [11–19] . Specifically , mothers harboring loss of function mutations in ZnT2 produce breast milk containing very low zinc levels; consequently , their exclusively breastfed infants suffer from severe zinc deficiency . These TNZD infants present with dermatitis , diarrhea , alopecia , loss of appetite , and consequently display growth and developmental delays [9] . Importantly , the unaffected healthy ZnT2 allele in TNZD was recently shown to be insufficient to provide the necessary high levels of zinc in breast milk , which are strictly required for proper infant growth and development [9 , 16] . Currently , no zinc transporter other than ZnT2 is known to play such a vital role in zinc transport into human breast milk . Taking into consideration the crucial role of ZnT2 in human health , we undertook the current study to understand the molecular mechanism underlying transmembrane zinc transport through ZnT2 . As a step towards this end , we used multiscale computational analyses and structural modeling to delineate the zinc permeation pathway and study the conformational dynamics of the transporter . We then functionally validated our proposed permeation pathway using site-directed mutagenesis and experimental zinc transport assays . To date , there is no high-resolution structure of the clinically significant ZnT2 transporter . However , E . coli’s YiiP , the closest ZnT2 homologue with a known crystal structure , is assumed to harbor a similar 3D structure [13] . YiiP and ZnT2 share ~20% sequence identity ( ~28% similarity ) along the region aligned ( ZnT2 residues 70–372 ) , allowing homology-based 3D model reconstruction of ZnT2 . Interestingly , YiiP was recently suggested to be involved in antibiotic resistance in Pseudomonas Aeruginosa [20] which enhances the importance of identification of the ion permeation pathway of YiiP as well . Herein , we conducted an array of calculations on the available structures of the bacterial YiiP ( X-ray and cryo-EM ) , and the 3D model of the human ZnT2 both in the inward- and outward-facing conformations . We investigated the zinc permeation pathway with different multiscale strategies , ranging from PDLD/S-LRA binding free energy calculations , CG evaluation of conformational change process and Monte Carlo simulations . It should be noted , with respect to zinc binding , that the large experimental free-energy of solvation for a zinc ion ( -467 kcal/mol [21 , 22] ) should be compensated by interactions with the transporter residues , and this is a computationally challenging task which was handled in our case by using explicit ligand particles ( see below as well as the Methods section ) . We complemented our computational work with site-directed mutagenesis , subcellular localization and functional zinc transport assay to delineate the putative zinc permeation pathway , highlighting key residues predicted to facilitate transmembrane zinc ion translocation through ZnT2 . The current study demonstrates the strengths of multiscale modeling and highlights the benefits of combining computational and experimental approaches to address medically important questions . We present the first proposed zinc permeation pathway of a human zinc transporter , bearing important implications for pathophysiological states of zinc deficiency and possible development of proper therapeutic interventions for zinc deficiency-associated disorders .
The ZnT2 models were constructed using the X-ray structure of the outward-facing ( OF; [23] ) and the cryo-EM structure of the inward-facing ( IF; [24] ) conformations of YiiP ( Fig 1 ) , based on their respective sequence alignments by satisfaction of spatial constraints ( S1 Fig ) using the Memoir [25] modeling suite . Memoir is specifically designed for transmembrane proteins and performs better than HHpred [26] and Swiss-Model [27] . The quality of the models was verified using Verify3D , providing a global and local assessment of the correctness of fold structures at the residue level . We found that Memoir models have better Verify3D scores than HHpred models and the scores were further improved after refinement with ModRefiner [28] . Accordingly , four systems constructed with Memoir and minimized with ModRefiner were chosen for the computations . We then minimized their energies and equilibrated them using the Molaris simulation package [29 , 30] . All systems were stable as manifested by low RMSD values to their initial coordinates ( less than 2Å for the backbone atoms ) . Following the construction and equilibration of these models and considering the expected homodimeric nature of ZnT2 [13 , 31] , we initially explored the possible location of the zinc permeation pathway along the dimerization interface , as previously suggested for YiiP [32] . However , a detailed examination of the structures and models revealed that there are almost no polar residues present along the dimeric interface , and in the OF conformation the monomers are not close enough to form a polar pore to stabilize zinc ions . In this respect , the bacterial Na+/H+ antiporter also functions as a dimer , whereas its sodium translocation pathway is located within each monomer and not between them [33–35] . We therefore searched for a highly polar zinc permeation pathway located within each YiiP/ZnT2 monomer and not between them . Consistent with previous studies , the X-ray structure of the bacterial YiiP monomer reveals a cavity leading from the zinc binding site to the extracellular milieu , lined with many polar and negatively-charged amino acid residues . However , the cavity grows very wide , and we therefore explored our models for several putative entry and exit routes and zinc permeation pathway . Prior to computing the binding energies of the zinc ion , we sought to study the protonation state for the binding site residues ( site A , see Fig 1 and [23] ) , harboring Asp and His residues . Determining the most stable protonation state is critical to attain correct binding energies of the ion because the charge distribution has arguably the biggest contribution to the interaction between the transporter protein and its zinc substrate . To that end , we calculated the relative energies of the different protonation states for YiiP and ZnT2 using the Molaris package ( see Methods for more details ) . The results are summarized in S2 Table , and the zinc binding calculations ( see below ) were conducted using the lowest-energy protonation state . As a control , we performed the binding calculations for several other protonation states for the OF YiiP structure and ZnT2 model , and the curves were very similar qualitatively , but showed a gradual difference quantitatively as the net negative charge increases , mostly around zinc binding site A ( S2 Fig ) . This hints at the interaction between protons and zinc ions , where protonation of side-chains weakens the binding of zinc ions , as expected for a putative proton-coupled zinc exchanger . Further investigation of the coupling between the protonation state and the translocation of zinc ions is beyond the scope of this work and will be explored thoroughly in a subsequent study . The zinc binding energy profile for the most stable protonation states ( see above ) on the equilibrated structures of ZnT2 and YiiP are depicted in Fig 2A . The binding energies were calculated using the Molaris PDLD/S-LRA method developed and refined over the years by the Warshel group [36 , 37] and proven valuable in numerous studies starting over 35 years ago , exploring various and diverse biological systems [29 , 37 , 38] . The curves were produced by averaging over several pathways selected along the cavities , where the MD simulations allow extensive sampling for the zinc ion position as well as the local conformation of the transporter protein ( see the Methods section for more details ) . We computationally explored several alternative permeation pathways based on the volume available in the open-direction cavity ( outward in the OF conformation and inward in the IF conformation , converging at site A ) . For the cavity on the closed side , we extrapolated the positions based on residues interacting with the zinc ion in the opposite conformation , as well as inspecting the conformational change in the CG and morph trajectories ( see below ) . The energy curves of the OF conformation revealed a strong zinc-binding at the known binding site ( site A in Fig 3 and [23] ) , as would be expected for a zinc transporter ( corresponding to an affinity in the pM-nM range , see note regarding quantitative evaluation at the end of this subsection ) . Although there are no published experimental Kd values for zinc binding per se , in the study of Chao and Fu [39] on the E . coli zinc transporter ZitB ( 25% sequence identity to ZnT2 ) , zinc translocation is composed of a two-steps process: a relatively rapid binding of zinc followed by a rate-limiting step in which the transporter undergoes conformational changes . Our energy profiles suggest a qualitatively similar pattern . Their kinetic study on ZitB reveals KM values in the high μM range . However , this discrepancy can be attributed to different functions and properties of ZitB and ZnT2 , as well as possible deviations between the value of KM and Kd . In this respect , the abundant intracellular zinc binding protein metallothionein ( capable of transferring zinc to the apo-forms of zinc-dependent enzymes and presumably to zinc transporters as well ) displays a KZn of 3 . 2x1013 M-1 ( i . e . a high metal binding constant ) , hence being consistent with the predicted concentration range [40 , 41] . Moving further along the energy curves , from the binding site to the open side direction , the binding energy steadily increases , albeit with relatively small barriers , as the interactions between the binding site residues and zinc are weakening , until a plateau is observed towards the bulk . To ensure that the suggested permeation pathway allows for selective zinc transport via alternating access , we additionally computed the energy profile in the closed direction . Indeed , Fig 2A shows high energy barriers for ions on the closed side of the transporter , compared to the open side , typical for alternating access . We then repeated this process for the IF conformations . Here we qualitatively obtained curves similar to the ones for the OF conformations , but as mirror images . The selection of the permeation pathway was not as obvious as above , because the cavity was not as large and visually apparent . Thus , we generated several putative trajectories for the conformational transition from OF to IF and examined very carefully the changes between one conformation and the other , i . e . where do these cavities form , and which interactions are disrupted . We produced the trajectory using two strategies: ( i ) our newly developed CG normal mode MC simulation ( see the Methods section ) ; and ( ii ) a Cartesian morph . Although these trajectories are not guaranteed to represent the precise physiological conformational change , we found them very instructive and were able to suggest several pathways , albeit more tightly packed than in the OF paths . Our energy calculations support the suggested permeation pathways , as they represent expected curves for zinc ion translocation across a transporter: low energies for the binding site , with barriers on both sides and a higher barrier on the closed side ( Fig 2A ) . In Fig 2B we show the energy landscape for the zinc ion as a function of its position within the protein and the ZnT2 conformation ( using the OF and IF end points ) . We estimated the conformation change barrier roughly at ~16 kcal/mol based on the ~1 . 5 sec-1 rates reported for YiiP [42] . Walking along the energy landscape ( Fig 2B and S1 Movie ) , we divided the path into three sections for clarity; the zinc ion enters the IF conformation from the cytoplasm and binds at the binding site ( site A [23]; section 1 ) ; then , ZnT2 undergoes a conformational change to the OF state ( section 2 ) , and finally the zinc ion exits to the vesicular lumen ( for ZnT2 ) or to the extracellular milieu ( for YiiP; section 3 ) . In this model , the transport of the zinc ion along the membrane necessitates a conformational change of the transporter protein , since the IF or the OF conformations alone harbor a high energy barrier . As mentioned above , coupling zinc translocation to the proton gradient and the directionality will be explored in a subsequent dedicated study . Since the energy profiles are a crucial component of the current study , we sought to assess their convergence . To validate the robustness of our methods and models and to prove proper sampling for the position of the ion and the local conformation of the interacting protein residues at each point , we chose the ZnT2 OF system as a control , and performed the same calculations using 10-fold longer simulations . The results obtained were essentially the same regardless of simulation length ( see S3 Fig ) , indicating that our calculations most probably converged within the original simulation times used . We wish to emphasize that the validity of our calculated permeation pathway is reinforced by the experimental mutational analysis and their consequences presented below . Therefore , the precise qualitative nature of the binding curves found in the current study ( e . g . the Kd we obtained and the conformational change energy ) are prone to uncertainty resulting mainly from the modeling process as well as limitations of the computational methods ( e . g . simulating a single monomer , not considering the probability average of all the different protonation states that are slightly higher in energy ) . Thus , the computational results and conclusions of this work are likely to be correct , while the calculated values should be still considered as a qualitative trend rather than actual quantitative numbers . The consistency of the calculated results with the observed mutational experimental analyses below supports the acceptance of this mode of calculation as a proper representation of the functional key residues along the putative zinc permeation pathway . To provide functional validation to our proposed zinc permeation pathway , we next aimed at predicting the amino acid residues that might play a functional role in zinc permeation . We searched for residues that contributed significantly to the calculated binding free energy ( see Fig 2 ) along the permeation pathway , in several trajectories , and in several positions . We note that considering stabilizing interactions at the energy barriers is important as well , because they lower the barrier on the open-side of the transporter . Notably , one of the binding sites revealed by Lu et al . , ( site B; [23] ) does not appear to directly participate in the zinc permeation pathway . Although this site appears to be sufficiently close to the permeation pathway to participate , a careful examination of the CG and morph trajectories strongly indicates that this is not the case . Site B is deformed in the IF conformation , and its residues are too far from the putative permeation pathway that we suggest . In this context , we propose that site B may act as a ‘waiting-area’ for zinc ions; that is , zinc ions initially bind to site B and are then translocated to the zinc permeation pathway . This might prove functionally crucial since zinc is found at very low intracellular concentrations , and it is bound and shuttled by zinc-binding proteins such as metallothioneins [9 , 10] . Thus , based on our computations and models we hypothesize that an auxiliary binding site to capture zinc when the transporter is undergoing conformational changes ( i . e . during the transport cycle ) might render a more continuous and robust ion flux . This too will be further investigated in dedicated future studies . Interestingly , according to ZnT2 modeling , the residues that participate in zinc ion binding ( site A ) are very similar in ZnT2 ( two Asp and two His ) and in YiiP ( three Asp and one His , see S1 Fig ) . In support of this suggestion , Hoch et al . , previously showed [42] that these differences between YiiP and ZnT5 or ZnT8 sequences contribute to the selectivity of the transporters towards zinc as a substrate , compared to YiiP which transports cadmium and zinc [42] . After careful consideration of the residues interacting with the zinc ion along the putative permeation pathway , we considered their evolutionary conservation ( see S1 Fig ) and assembled a list of candidate residues that are both important for zinc interaction and hydration , and are also conserved . This was done following the rationale that evolutionary-conserved residues are more likely to be functionally important . Herein , the residues studied by site-directed mutagenesis , shown in Fig 3 , were carefully selected according to their evolutionary conservation and their calculated energy contribution to zinc binding . To provide experimental validation for the computational analyses delineating the putative zinc permeation pathway of ZnT2 , we mutated selected residues and assessed their impact on actual zinc transport activity in live human cells . While employing Ruby-tagged ZnT2 expression plasmids ( emitting red fluorescence ) , the green fluorescence of the selective zinc probe FluoZin 3-AM , was used . Hence , zinc-containing vesicles were detected as green fluorescent vesicles solely in cells transiently transfected with an active ZnT2 transporter after incubation with zinc as previously described [19] ( Fig 4 ) . Our choice of specific protein residues to study was made by energetic , structural , and evolutionary conservation basis , as explained above . Fig 5B shows a ZnT2 monomer color coded by conservation from the extracellular side of the membrane . Only residues located at the transmembrane domain and with the highest degree of conservation ( 9 ) and the lowest ( 1 and 2 ) are shown for clarity . This panel shows that the vast majority of highly conserved residues face the internal pore of the transporter , while the fast-evolving residues are facing the lipid core of the membrane . This fits the model fold and the expected biological function of ZnT2 , where functional conserved residues lining the permeation pathway cannot tolerate evolutionary changes , while more rapidly evolving hydrophobic residues facing the membrane are able to accommodate a range of hydrophobic residues without impacting the membrane-protein interaction . Therefore , the positions we chose have an important energetic contribution to zinc binding , are highly conserved , face the pore according to our structural model , and in consequence , are expected to impact zinc binding . The results of the site-directed mutagenesis are summarized in Table 1 . Additionally , we computed the evolutionary conservation of the ZnT2 family . Fig 5 shows that most mutated positions are highly conserved ( in purple ) and face the putative zinc permeation pore ( Fig 5 ) , with the exception of H197 and Q198 which are slightly variable . All mutants displayed similar or even higher red fluorescence levels ( Ruby fluorescence ) compared to the WT-ZnT2 , indicating the proper expression of these ZnT2 mutants ( Table 1 , right column ) . For our analysis , we grouped mutated positions according to their physicochemical , biological and evolutionary conservation relevance , and hence in this way their biological impact and phenotypic characteristics could be more easily understood ( notice that the groups are not necessarily exclusive ) . The first group includes residues E88 , D103 and E140 , which are likely involved in direct interaction with zinc; these residues are negatively charged , highly conserved and located along the putative zinc permeation pathway according to the proposed 3D model . Indeed , site-directed substitution of these residues to Ala proved to be highly deleterious for ZnT2 function with about 90% loss of WT ZnT2 zinc transport activity ( Table 1 ) . Thus , we considered these positions to be important for zinc transport . Notably , whereas the E140A mutant showed a low but significant 20% decrease in the overall number of ZnT2 vesicles when compared to the WT ZnT2 , the decrease in its zinc transport function was much more profound . The second group includes residues M85 , M114 , N189 and N214; these residues are conserved and are spatially located below and above the zinc-binding site with respect to the membrane plane ( Fig 5 ) . However , in contrast to the first group , this group of residues showed a moderate impact on zinc transport , upon substitution to Ala , with 24–52% decrease in zinc transport capacity ( Table 1 and Fig 4 ) . This suggests that the impact of polar residues along the permeation pathway on binding and/or transport of zinc is manifested collectively , as the contribution of each single residue to zinc permeation is smaller compared to the charged residues aforementioned . We also observed an additional impact on ZnT2 function; for example , E88A and N189A displayed loss of vesicular localization ( Fig 4 ) and furthermore , M114A , E140A and N214A exhibited a decreased number of vesicles per cell when compared to the WT-ZnT2 ( Table 1 ) . Supporting our findings and applying a different experimental setup , a recent study on the closely related ZnT1 revealed that ZnT1’s equivalent of ZnT2’s mutants , E88A and N189A , rendered ZnT1 dysfunctional [43] . Our findings suggest that these five residues ( E88 , M114 , E140 , N189 , and N214 ) have a significant role in protein structure and stability in addition to their role in zinc binding and zinc permeation . In contrast , D103A displayed a high number of ZnT2 vesicles while showing very little zinc accumulation , suggesting that this residue has an important role in zinc transport with little impact on transporter stability and subcellular localization . The third group of mutants including M85A , H197A and Q198A retained the canonical subcellular vesicular localization ( Fig 4 and Table 1 ) . In this respect , H197 and Q198 are not evolutionary conserved residues , are located near the cytoplasmic region , and point away from the zinc permeation pathway ( Fig 5B ) , according to the proposed structural model . Hence , they were not expected a priori to impair zinc transport . Indeed , H197A and Q198A retained 80–90% of WT-ZnT2 zinc transport activity , with no significant differences when compared to zinc accumulation mediated by WT ZnT2 ( Fig 4 and Table 1 ) . We also focused on a fourth group including residues H201 , H203 , and H205 , which are part of the conserved GHGHSH motif ( His-rich loop ) located between TM helices IV and V in ZnT2 but are not present in the bacterial homologue , YiiP . This motif was previously suggested to be involved in sensing cytosolic zinc levels [44 , 45] or in mediating the activation of tissue-nonspecific alkaline phosphatase ( TNAP ) by ZnT5 [46] . However , based on our model of the human ZnT2 , all three His residues are pointing away from the central permeation pore and are positioned in close proximity to the exit of the putative permeation tunnel . In different initial structural models obtained for ZnT2 , these three His residues were in very different conformations , since the GHGHSH motif is a highly charged flexible loop and lacks complete template information ( see S1 Fig ) , thus had to be modeled ab initio by FREAD in Memoir suite . In most models , these three His residues face the water bulk or other peripheral residues and their high conservation and positive charge suggests a role in the stabilization of ZnT2 with acidic phospholipid head groups or an allosteric zinc regulation role , rather than direct interaction with the transported zinc ion . Indeed , substitution of all three His residues to Gly did not exert any deleterious effect on the zinc transport capacity of ZnT2 . This further suggests that the GHGHSH motif in ZnT2 is not directly involved in zinc translocation to site A and across the transporter , i . e . there are no direct interactions between the zinc ion ( or its first hydration shell water molecules ) and the histidine side chains of the GHGHSH motif . A very recent publication strengthen our findings , showing that deletion or substitution of these His residues to Ala in ZnT2 did not affect zinc transport activity [71] . However , an allosteric regulatory role of this motif cannot be excluded . In summary , site-directed mutagenesis of seven key residues along the putative zinc permeation pathway of ZnT2 , markedly impaired its zinc transport function . In contrast , site-directed mutations at non-conserved residues ( H197A and Q198A ) , and at the conserved GHGHSH motif that is suggested to point away from the zinc permeation pathway , had only a minor deleterious impact on the zinc transport function of ZnT2 . Taken together , residues predicted by our model and free-energy calculations , supported by the conservation analysis , agree well with our experimental validation of ZnT2 functionality .
In this study we demonstrated the power and potential of multiscale approaches by analyzing the zinc permeation pathway of ZnT2 , highlighting the constructive synergism of computations and functional validation experiments . The principles of multiscale modeling used in this study , include representing the model structure using a CG model with a reduced number of atoms . Such approaches are highly advisable to accelerate a simulation ( by reducing the degrees of freedom ) and to perform calculations that would otherwise be extremely challenging ( such as NMA ) . The binding free-energy calculations were also performed with a semi-macroscopic method , which is a form of reduced-dimensionality modeling , allowing faster convergence and thus more stable results . The reader is directed to other studies presented in this special issue , regarding other various multiscale approaches and techniques . More specifically , in the current study we undertook extensive CG simulations of our proposed model of ZnT2 , which provided structural and evolutionary information which delineates , for the first time , the putative zinc permeation pathway of ZnT2 , from the cytoplasm into the lumen of intracellular vesicles ( or the extracellular milieu in the case of YiiP ) . This proposed permeation pathway harbors one central zinc binding site ( site A [23] and Fig 3 ) and two cavities showing alternating-access in the two principal conformations of YiiP and ZnT2 and possibly in other zinc transporter homologues as well ( Fig 2 ) . To functionally validate this permeation pathway experimentally , we performed site-directed mutagenesis to target various residues along the zinc permeation pathway ( Fig 4 ) . Our rationale was to try to properly predict , based on the YiiP crystal structure and homology-based model of ZnT2 , the correct deleterious impact of site-directed mutagenesis of key residues along the permeation pathway on the zinc transport activity of ZnT2 in viable cells . Indeed , this process lent strong experimental support to our structure modeling and zinc permeation pathway prediction . Interestingly , a recent cryo-EM structure of the ZnTB zinc transporter was revealed [47] , showing a pentameric architecture . ZnTB is proton-driven , precisely as ZnT2 is considered to be , however their sequence similarity is very low , and therefore ZnTB was not deemed a suitable template for ZnT2 in our present work . We selected residues facing towards the putative permeation pathway with high contribution to the calculated zinc binding energy and used conservation analysis as a cross validation prior to the mutagenesis . Indeed , all the conserved residues that we computationally predicted to be important for zinc permeation , experimentally impaired the zinc transport function of ZnT2 upon substitution to alanine , while five of these mutants exhibited proper vesicular localization . As a negative control , we showed that residues that were facing away from the zinc permeation pathway including H197A , Q198A , and the GHGHSH motif , or having a tendency towards a less organized secondary structure ( i . e . disordered region ) , showed a minimal deleterious effect on zinc transport activity of ZnT2 , hence further supporting our hypothesis . Thus , multiscale computational analyses complemented with functional experimental validation lead to the construction of valid models for the 3D functional conformations of human ZnT2 and the zinc translocation pathway . Furthermore , we were able to successfully predict important residues around the putative binding site involved in zinc binding and translocation , both in the direction of the cytoplasm and the extracellular milieu . From a translational medicine perspective , such zinc permeation pathway studies may facilitate the screening and identification of small molecules that can correct the proper folding and/or function of mutant transporter proteins . In this respect , pharmacoperones are recently emerging as a novel class of hydrophobic small molecules that can bind to mutant misfolded and inactive proteins , thereby restoring their proper folding , subcellular sorting , and function [48 , 49]; this novel approach is currently known as pharmacoperone drug therapy . These mutant proteins which are associated with various human disorders include enzymes , receptors , channels , and transporters [48 , 49] . For example , Menkes disease is a neurodegenerative disease presenting with seizures , lethargy and hypotonia , resulting in death in early childhood . This disorder is caused by copper deficiency due to mutations in ATP7A that encodes for the copper-transporting ATPase 1 transporter [50] . Copper-transporting ATPase 1 , which is localized in the trans-Golgi network , functions as an ATP-driven intracellular pump that transports copper into the trans-Golgi network for incorporation into copper-requiring enzymes; whereas , those pumps that are localized on the plasma membrane are involved in copper transport out of the cell [51] . Some of the mutant copper-transporting ATPase 1 transporters are misfolded and hence retained in the ER [50] . These mutants can be corrected by pharmacoperones , further supporting the suggestion that the ER-retained mutant is misfolded [50] . In vitro , copper supplementation of the culture media results in the correction of the mislocalization [50] . Therefore , the substrate for the transporter serves as a pharmacoperone , facilitating the folding of the mutant protein . In Menkes patients with residual copper-transport activity , early treatment with copper injections can normalize clinical outcomes and increases survival [52] . Hence , it is possible that such pharmacoperones could be identified for mutant zinc transporters like ZnT2 and other zinc transporters including ZIP4 that can correct the proper folding , subcellular sorting , and zinc transport , thereby providing a targeted therapeutic strategy to overcome well-defined diseases associated with zinc deficiency or other solute transport deficiencies . We hope that our present study will enhance our understanding of zinc transporters towards this strategy , and believe that the approach presented herein will convince others to consider multiscale modeling and experimental complementation as a platform to address pressing biomedical issues .
ZnT2 models were generated based on the OF and IF structures of E . coli YiiP ( PDB: 3H90 and 5VRF , respectively ) . Modeling of ZnT2 was carried out by Memoir method [25] , specifically designed to exploit the structural constraints imposed by membrane proteins . Briefly , homologous sequences are aligned by MP-T [53] , guided by the membrane information from iMembrane [54] . Membrane information is again used in model building by the MEDELLER program , designed specifically to construct homology-based membrane protein core structures [55] . Finally , the model built by MEDELLER is completed with a membrane protein-specific version of the FREAD loop-modelling method [56] . All homology models were refined and minimized by ModRefiner [28] . All models had similar Verify3D profile average scores [28] . Alignment of the ZnT2 model to the YiiP structures used one point per residue ( the CA atom ) . For visualization purposes , Fig 3 fitting was performed using only the C-terminal domain residues , whereas Fig 5 fitting was performed using all residues . The systems for YiiP and ZnT2 , both in the IF and OF conformations , were constructed as follows: the transporter homodimer was inserted into a 30Å-thick 3D particle grid emulating a hydrophobic membrane , using the Molaris software package [29 , 30] . The gap between the protomers was treated as part of the membrane milieu , as this gap exists within the expected membrane space and it is suggested to be non-functional [24] . Then , the systems were hydrated with a 40Å-radius water sphere , using explicit 3-particle water molecules , and were submitted to energy minimization using the steepest descent algorithm followed by a short local relaxation simulation of 100 ps . Relaxation was performed in the presence of a zinc ion at the putative binding site ( based on the structures of 3H90 and 5VRF ) , to prevent charge repulsion between the binding site residues . To obtain the binding energy curves , the zinc ion at site A was vertically moved by 1Å intervals at the z-axis , to either side of the membrane , until it reached a pre-selected point in the water , i . e . the extracellular ( vesicular lumen ) or intracellular bulk . For each such position of zinc , 10 PDLD/S-LRA calculations ( see below ) were performed entailing an additional short step of relaxation of 10–20 ps , allowing the transporter to relax around the new position of the zinc ion , and to generate different starting configurations . Being a semi-macroscopic method , with CG elements , the PDLD/S-LRA method converges substantially faster than umbrella sampling/PMF methods , and calculations using typically these parameters have been successfully applied to many complex biological systems ( e . g . [3 , 5 , 29] ) , regardless , we performed longer simulations as a control for one example ( see Results and S3 Fig ) . During the relaxation process , the zinc ion could freely move in the xy-plane . This treatment reduced as much as possible the bias of selecting the path manually . For the open direction , where the cavity was very wide , we selected 5–7 such trajectories ( i . e . different final points in the bulk ) , whereas only 3–4 trajectories were selected in the closed direction . Along the trajectory in the closed direction we computed short distances between the zinc ion and the transporter that are most probably non-physiological . These tight spots result in high-energy barriers that suggest an ion does not bind at this position in the biological system , but the energy could still be estimated and reported using the computational methods described below . The binding energy calculation for each position point was performed using the scaled semi-macroscopic Protein Dipoles Langevin Dipoles approach ( PDLD/S ) of Molaris [29] . Water in this method is represented semi-macroscopically by Langevin dipoles . The energy is the average of the charged and uncharged states , following the linear response approximation ( LRA ) , scaled using a dielectric constant ε = 8 for the protein . Convergence was achieved by running molecular dynamics ( MD ) simulations for the relaxation and averaging the results of the different conformations ( PDLD/S-LRA ) [36] . The reported energy was the average of all repeats ( n = 10 ) and for all trajectories ( 3–7; see above ) . Additionally , running averages were used in the full-atomistic PDLD binding curves , applying a window size of three , to better match the CG-nature of the system used for the conformational changes . The calculations were performed for several protonation states but only one protonation state is discussed in the main text based on the relative energies of the states ( see below and see S2 Fig ) . YiiP and ZnT2 have 4 ionizable residues in binding site A ( DDHD and HDHD , respectively ) . To assess which protonation state is the lowest in energy , we computed the total energy of the electrostatic cluster following the formalism presented previously [37 , 57] . The total energy of each state is given by the sum of: ( i ) the solvation energy of the ionized residues ( representing the energy cost of bringing the ionized residues and the zinc ion from the bulk to the interior of the protein , relative to the system with zero charges ) ; ( ii ) the energy of ionizing the given residues ( His or Asp ) in water , based on bulk pKa values; ( iii ) the electrostatic energy between the ionized residues ( Coulombic energy to bring the ionized residues from an infinite distance to their distance calculated in the binding site , using a dielectric of 80 ) ; and ( iv ) the electrostatic energy between the ionized residues and the zinc ion . We include the zinc ion to determine the protonation state energy because the states are used to compute the zinc binding energy curves . Therefore , we need to consider the most stable protonation state in the presence of zinc . S2 Table lists all the protonation states and associated values . The zinc ion was simulated using three different settings: ( i ) a single charged particle; ( ii ) a 7-particle entity arranged as an octahedral complex; ( iii ) and a 5-particle entity arranged as a tetrahedral complex , based on ligand-field theory [58] . Previous studies from the Warshel group showed that transition metals are better simulated when they are represented by an ion-and-ligands structure rather than a point charge . The partial charges as well as radii of the particles in the different models were calibrated such that the hydration energy of the zinc ion in water would match the literature value of -467 kcal/mol previously reported [21 , 22] . These settings in the protein yielded very similar results , thus we ultimately used the 5-particle entity for the reported energies in the present study . For the conformational change MC trajectory , we used our newly developed method that will be described extensively in a future publication . In brief , we look for conformational transitions in the landscape of our simplified CG model [59 , 60] , where the water is treated implicitly . To do so we consider first only the non-bonding and bonding interactions , evaluate the Cartesian second derivative and evaluate the corresponding Cartesian normal modes , which are then projected on the torsional angle ( dihedral ) space . Then we move along the torsional normal modes and generate a path from the initial to the final structure ( in this case , the IF and OF conformations ) , using a Monte Carlo ( MC ) procedure with the Metropolis acceptance criterion . Finally , we evaluate the full CG energy along the MC generated path . MCF-7 breast cancer cells were grown and transiently transfected with pcDNA3 . 1 zeo expression plasmids harboring WT ZnT2 or ZnT2 mutants tagged with a Ruby fluorescent protein as previously described [19 , 61] . Primers used for site-direct mutagenesis are listed in S1 Table . The empty vector of RFP was used as a negative control for zinc transport . Eighteen hours after transfection , cells were incubated for 1 hour in growth medium containing 75 μM ZnSO4 and then stained with a selective fluorescent zinc probe FluoZin3-AM ( 1μM ) . Cells were then analyzed using Aria IIIu-flow cytometer ( for measuring median FluoZin3-AM and Ruby/RFP fluorescence ) , IN Cell Analyzer 2000 ( for quantification of the number of vesicles per cell ) , or imaged by confocal microscopy as previously described [16] . Percentage of FluoZin3-AM fluorescence was calculated by dividing the median fluorescence intensity of FluoZin3-AM , as measured in the cells after transfection with the different mutant ZnT2 plasmids , by the median fluorescence intensity values of the WT ZnT2 protein . These values reflect the zinc transport activity of the different ZnT2 mutants compared to the WT ZnT2 ( Table 1 ) . To test the hypothesis that zinc transport is lower in cells transfected with mutants ZnT2 as compared to cells transfected with WT ZnT2 , we compared the median FluoZin3 fluorescence levels in cells transfected with ZnT2 mutants to the WT ZnT2 , using one tailed T-Test with unequal variance . The hypothesis testing was followed by False Discovery Rate correction for multiple hypothesis testing with α = 0 . 05 [62] . The same statistical analysis was performed to test the hypothesis that Ruby fluorescence ( indicating ZnT2 expression levels ) is lower in cells transfected with ZnT2 mutants as compared to cells transfected with the WT ZnT2 . Table 1 shows the fluorescence intensity values as percent of WT for clarity; statistical analysis was conducted on the actual fluorescence levels of each mutant compared to the WT fluorescence . Estimation of the evolutionary conservation for each protein position was computed using the ConSurf server [63–66] . Performed with the UCSF Chimera package [67] , VMD [68] , and PyMOL [69] . | Herein we employed multiscale modeling and electrostatic energy calculations to delineate , for the first time , a putative zinc permeation pathway , from the cytoplasm into intracellular vesicles ( for ZnT2 ) or to the extracellular milieu ( for YiiP ) , along the membrane-spanning domain of the human zinc transporter ZnT2 and its E . coli homologue , YiiP . These computational findings were functionally validated using site-directed mutagenesis of ZnT2 residues predicted to reside along the putative zinc permeation pathway and zinc transport assay . Our results shed light on the transport mechanisms of ZnT2 and YiiP and pave the way towards the elucidation of the zinc translocation mechanism in other ZnT family members . Furthermore , these findings could also be harnessed to the possible development of therapeutic interventions in zinc-associated pathologies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"vesicles",
"zinc",
"transporters",
"built",
"structures",
"engineering",
"and",
"technology",
"site-directed",
"mutagenesis",
"simulation",
"and",
"modeling",
"nutrition",
"biological",
"transport",
"molecular",
"biology",
"techni... | 2018 | Demonstrating aspects of multiscale modeling by studying the permeation pathway of the human ZnT2 zinc transporter |
How do humans and other animals face novel problems for which predefined solutions are not available ? Human problem solving links to flexible reasoning and inference rather than to slow trial-and-error learning . It has received considerable attention since the early days of cognitive science , giving rise to well known cognitive architectures such as SOAR and ACT-R , but its computational and brain mechanisms remain incompletely known . Furthermore , it is still unclear whether problem solving is a “specialized” domain or module of cognition , in the sense that it requires computations that are fundamentally different from those supporting perception and action systems . Here we advance a novel view of human problem solving as probabilistic inference with subgoaling . In this perspective , key insights from cognitive architectures are retained such as the importance of using subgoals to split problems into subproblems . However , here the underlying computations use probabilistic inference methods analogous to those that are increasingly popular in the study of perception and action systems . To test our model we focus on the widely used Tower of Hanoi ( ToH ) task , and show that our proposed method can reproduce characteristic idiosyncrasies of human problem solvers: their sensitivity to the “community structure” of the ToH and their difficulties in executing so-called “counterintuitive” movements . Our analysis reveals that subgoals have two key roles in probabilistic inference and problem solving . First , prior beliefs on ( likely ) useful subgoals carve the problem space and define an implicit metric for the problem at hand—a metric to which humans are sensitive . Second , subgoals are used as waypoints in the probabilistic problem solving inference and permit to find effective solutions that , when unavailable , lead to problem solving deficits . Our study thus suggests that a probabilistic inference scheme enhanced with subgoals provides a comprehensive framework to study problem solving and its deficits .
Problem solving consists in finding efficient solutions to novel tasks for which predefined solutions are not available [1] . Humans and other animals can efficiently solve complex problems [2 , 3] but the underlying neuronal and computational principles are incompletely known . Research on the neuronal underpinnings of problem solving has often proceeded in two different ways . First , researchers have focused on how individual brain areas or circuits solve problems in specific domains; for example , the hippocampus is considered to be implied in solving navigation problems [4–6] and parieto-frontal regions are considered to be implied in mathematical problem solving [7] . This approach is compatible with the idea that the brain has dedicated neuronal machinery to solve domain-specific problems , with little hope to find common principles across them . A second line of research has focused on domain-general problem solving strategies , as exemplified in the realization of general problem solvers and other influential cognitive architectures in cognitive science [1 , 8–13] , planners and problem solvers in AI [14–16] , and the recent view of the brain as a statistical engine [17–19] . A challenge in this second research line is to identify core computational principles of planning and problem solving that are , on the one hand , valid across multiple cognitive domains ( e . g . , sensorimotor tasks , navigation , and mathematical problem solving ) and , on the other hand , can be implemented in neuronal hardware and work well in ecologically valid contexts [20] . In this article we show that problem solving can be characterized within a probabilistic inference framework . This framework is increasingly used across multiple domains ( sensorimotor [21 , 22] , decision-making and planning [23–25] , human-level reasoning [26–28] and learning [29] ) and levels of description ( higher / computational and lower / neuronal [17 , 18 , 30–33] ) , supporting the idea that problem solving does not necessarily require specialized mechanisms that are distinct from those used by perception and action systems . Our problem solving approach is framed within the planning-as-inference ( PAI ) framework , which casts planning as a probabilistic inference problem [23 , 34–38] . In this perspective , goals are “clamped” ( i . e . , they are treated as “future observations” that the system strives to achieve ) and probabilistic inference permits to select the sequence of actions that fills the gap between current and goal states . Despite its usefulness to explain goal-directed behavior [25 , 39–41] and to design robot architectures [42] , the standard PAI framework fails to capture some important aspects of ( human ) problem solving , such as the ability to exploit the “junctions” of problems and to subdivide them into more manageable subproblems . Here , in keeping with a long tradition in human problem solving and cognitive architectures , we augment the PAI approach with a subgoaling mechanism that permits splitting the original problem into more manageable , smaller tasks whose achievement corresponds to milestones or subgoals of the original problem [1] . For example , navigation problems can be decomposed by using subgoals ( e . g . , landmarks ) such as “reach the Colosseum , then reach the Imperial Forum” ( if one lives in Rome ) and puzzles like the Tower of Hanoi can be decomposed by using subgoals such as “free up third rod” . The importance of subgoaling has been widely recognized by the most popular architectures for symbolic problem solving [1 , 8 , 10 , 13] and in other domains such as connectionist networks , hierarchical reinforcement learning , AI , planning , and robotics [14–16 , 35 , 43–48] . However , opinions differ about the mechanisms underlying subgoaling . Most systems , especially within the human problem solving and AI traditions [1 , 13] , assume that subgoaling proceeds backward from the final goal states and serves to resolve “impasses”: if a goal-achieving action cannot be executed because of a missing precondition , achieving the precondition becomes the next system subgoal; and so on . Instead , converging evidence from empirical studies highlights the importance of feedforward mechanisms for subgoaling and “search” ( i . e . , mechanisms that proceed from the current to the goal state ) in living organisms . The importance of feedforward mechanisms emerges from psychological experiments [49 , 50] as well as neurophysiological studies on rodents [51–53] , monkeys , [54 , 55] and humans [56] . In keeping , we implement subgoaling within a feedforward probabilistic inference ( PAI ) scheme , showing that the proposed method reproduces characteristic signatures of human problem solving . To this aim , we present four simulation experiments in the widely-used Tower of Hanoi ( ToH ) task [57] . We show that in our simulations successes and failures of solving the ToH ( e . g . , the failure of dealing with counterintuitive movements ) correspond to the successful identification or a misidentification of subgoals during the inference , respectively .
The Tower of Hanoi ( ToH ) task has been widely used in neuropsychology to study executive function and deficits in planning [57 , 58] . A standard ToH consists of three disks having different sizes that can slide into three rods to form goal configurations . The aim of the game is starting from any initial configuration of disks and reach a goal configuration using the smallest number of actions . Fig 1 shows sample initial ( a ) and goal ( b ) configurations of a ToH . The rules of the game prescribe that only one disk can be moved at a time and no disk may be placed on top of a smaller disk . The quality of the solutions found by subjects performing a ToH puzzle is usually described using the number of movements required to achieve the goal configuration and the reaction time [57] . To model the Tower of Hanoi task , here we use a standard approach in AI [59] that consists in mapping the original task into a path-planning problem , whose states and transitions are shown in Fig 1C . The resulting problem has 27 states ( squares ) and 39 transitions ( edges ) . To solve ToH problems ( e . g . , go from S27 to S20 ) , we use a feedforward probabilistic inference method that iteratively samples candidate subgoals from a prior distribution and uses them as waypoints until a complete solution is found . Specifically , our methods use and extend the planning as probabilistic inference ( PAI ) framework , in which planning is cast as a probabilistic inference problem [23 , 25 , 34 , 37 , 40 , 60] . In this perspective , a planning strategy can be attained by imposing ( “clamping” ) goal or rewarding states as ( future ) observations , so as to bias the probability of future events towards such desirable states . In problems like the ToH having moderately large state spaces , exploring all the possible paths is very demanding ( in some cases , computationally intractable ) . For example , despite the limited number of states and transitions of the Tower of Hanoi task shown in Fig 1C , this problem allows for about ∼1012 possible policies , i . e . , mappings from states to actions [61] . In this case , a mechanism for splitting the problem into more manageable subproblems—like subgoaling—is helpful . In keeping , our method differs from the standard PAI approach because we augment the probabilistic scheme with a subgoaling mechanism that permits to split the original problem into smaller ( and less complex ) subproblems . A second difference with the standard PAI approach is that ( akin to Active Inference theory [30] ) we do not treat goals as future observations but as states having high ( Bayesian ) priors . The rest of this Section introduces the key components of our approach . It is divided into four subsections . The first subsection introduces the probabilistic model ( Dynamic Bayesian Network [62] ) we used for the inference , which describes the conditional dependencies between the relevant stochastic variables ( e . g . , states , actions , and subgoals ) . Importantly , at difference with most planning-as-inference architectures , our model includes ( a probability distribution over ) subgoals . The second subsection thus illustrates two methods that we used to define the “priors” of such subgoal states . The former ( “algorithmic” ) method , which is based on calculability theory , constitutes a novelty of our approach . In the experimental section of this article , we use it to explain human performance in the ToH . The latter ( “perceptual” ) method incorporates a simpler , distance-based metric of the ToH problem space . In the experimental section of this article , we use it to explain failures in human problem solving ( e . g . , problems in executing counterintuitive movements ) . The last two subsections explain in detail the two nested procedures that compose the probabilistic inference: an “inner” procedure that produces candidate plans ( see Algorithm 1 ) and an “outer” procedure that implements a decision rule to select among them ( see Algorithm 2 ) . Essentially , in the “inner” procedure , several candidate plans are stochastically produced by our subgoaling-based probabilistic method and scored according to informational measures . In the “outer” procedure , a Bayesian voting procedure selects the best-posterior-valued candidate plan [63 , 64] . The probabilistic model used in the simulations is a Dynamic Bayesian Network ( DBN ) of Fig 2 . The nodes of the DBN are arranged on two layers corresponding to two consecutive slices of time indicated with subscripts , e . g . St and St+1 . First-order and stationary Markov properties hold: every variable depends exclusively on other variables expressed at the same or in the immediately preceding time step . The DBN model permits to formulate a problem solving task as a Markov Decision Process ( MDP ) , in which states and actions are described by the stochastic variables S and A , respectively . We assume that S varies in a discrete set consisting of integer values in the range {0 , … , n} , with n being the total number of states . The node A corresponds to seven different actions {act1 − act6 , ε}: move a disk from the first rod to the second rod ( act1 ) ; move a disk from the first rod to the third rod ( act2 ) ; move a disk from the second rod to the first rod ( act3 ) ; move a disk from the second rod to the third rod ( act4 ) ; move a disk from the third rod to the first rod ( act5 ) ; move a disk from the third rod to the second rod ( act6 ) ; and an auxiliary “rest” action ( ε ) , i . e , a transition from a state to the same state . Note that not all actions are defined in every state ( e . g . , in some states it is impossible to move a disk from the first rod to the second rod ) . The node Π in Fig 2 represents policies , or deterministic mappings from states to actions to be taken in these states . As a policy deterministically specifies an action for every state ( a = π ( s ) ) , executing the same policy for a number of steps determines a sequence of transitions amongst states [61] . The total number m of policies available depends on the number of states S and actions A in the environment . We include also a rest policy πε that associates the action ε to every state ( i . e . , ε = π ( s ) ) . Policy selection is modeled by a transition probability distribution p ( Π|s , sg ) , where s is one arbitrary state and sg represents one arbitrary subgoal . Potentially , every state s could be a subgoal and used for planning a strategy in order to achieve the final goal . For this reason , in our simulations the set of subgoals has the same cardinality as the set of states: {0 , … , n} . The final goal state is considered as a particular subgoal having the highest a priori probability . At each time step , a new state is determined based on the current state and the action given by the selected policy , according to p ( St+1|st , at ) . The subgoal transition follows instead the distribution p ( SGt+1|ft , sgt ) . The variable F monitors the agent’s progress in the task by reporting whether or not a goal or subgoal has been achieved , and determines when the inferential process terminates or a new subgoal needs to be sampled . F can only assume three discrete values: 0 , 1 , 2 , see [38] for a related method . It has value 2 if the agent has reached the final goal state ( in which case , the inferential process terminates ) . It has value 1 if the agent has just reached a subgoal state ( in which case , a new subgoal is sampled ) . Otherwise , F has value 0 ( in which case , the same subgoal is used for the next inferential step ) . Subgoals are states that are inferred during the planning-as-inference ( PAI ) process , and which enable the selection of optimal sequences of transitions from the initial state to the final goal . A key feature of our probabilistic model is the use of a subgoal a-priori probability distribution ( SG ) that is used to guide the inferential process and in particular to select candidate subgoals ( see later ) . For each state , this distribution essentially encodes a ( prior ) probability that the state is a potentially useful subgoal , or in other words a good way to carve the search problem . In the following , we show two ways to calculate the a-priori probability ( prior ) of subgoal states . The former method ( Algorithmic priors ) , crucial in our approach , is deduced from Algorithmic Probability theory , introduced by Solomonoff [65 , 66]; as we will discuss , it reveals structural aspects of the problem space and affords efficient path planning . The latter method ( Perceptual priors ) , carves the problem space in a different ( and suboptimal ) way , essentially encoding the mere perceptual distance to the goal state . We will use this latter method to simulate human failures in the execution of counterintuitive movements ( see the “Results” section ) . In the standard planning-as-inference ( PAI ) approach , the inference tries to find a suitable policy from a start ( s0 ) to a goal location ( sgoal ) . Rather , in keeping with the recognized importance of subgoaling in problem solving , our method infers a series of simpler sub-plans that pass through a sequence of subgoals s0 , sg1 , … , sgk , … , sgoal , until the goal state is reached . In other words , here the inferential process aims at finding a sequence of subgoals ( and associated policies that govern the transitions from one subgoal to the next ) that best permits to solve a specific problem , rather than finding a solution from start to end . The way the sequence of subgoals and associated policies are selected is described next . The procedure iteratively samples ( i . e . , extracts probabilistically ) , first , candidate subgoals from the previously described a priori subgoal distribution p ( SG ) , and second , policies ( from π ) that can govern the transition to the sampled subgoal . The sampling procedure is cycled until the final goal sgoal is eventually reached ( or for a maximum number of steps Tmax ) . During this procedure , both s and p ( SG ) are iteratively updated to reflect the current planned path toward the goal and the most useful subgoals given the final goal , respectively . In the course of this iterative process , the candidate subgoals are retained or discarded by considering the computational complexity of the sub-problems they generate . More formally , a candidate subgoal sequence is selected on the basis of the code length of the corresponding programs , which go from one subgoal ( or the start location ) to the next . We remind that a program is defined as the sequence of actions necessary for the transition from an initial state s to a subgoal state sg ( and is equivalent to a path following a policy π [61] from s to sg ) . The length of the program ( i . e . , the number of actions necessary to reach sg from s ) is converted into a probability using algorithmic probability theory [65 , 66 , 68 , 69] and this probability is used in the iterative procedure to decide whether the solution ( including the subgoal plus the program ) should be retained or discarded . The inference is formalized by the pseudocode of Algorithm 1 . Following the model of Fig 2 , the inference starts at time t = 0 from the initial state s0 of the node S . The distribution of the algorithmic priors p ( SG ) is modified to set the goal state sgoal as the state with highest prior . Algorithm 1 Planning Inference ( s0 , sgoal , q , p ( SG ) , Tmax ) Require: Starting state s0 , goal state sgoal , sampled model instances q , subgoal algorithmic priors p ( SG ) , maximum number of forward inferences Tmax . Ensure: State sequence [s0 , … , sgoal] , subgoal sequence Seq . 1: t = 0 2: set S0 to the starting state s0 3: sample a subgoal sg0 from the prior distribution p ( SG ) attained by using Eq ( 1 ) on each state 4: select a policy π0 maximizing Eq ( 8 ) sampled through a Monte Carlo method 5: determine the action a0 depending on π0 and s0 6: evaluate the termination condition state F0 according to p ( F0|sg0 , s0 ) 7: while ( Ft < 2 and t ≤ Tmax ) do 8: t = t+1 9: determine the state st by means of p ( St|a ( t − 1 ) , s ( t − 1 ) ) 10: select the subgoal sgt maximizing Eq ( 12 ) sampled through a Monte Carlo method 11: select a policy πt maximizing Eq ( 8 ) sampled through a Monte Carlo method 12: determine the action at depending on πt and st 13: evaluate the termination condition variable Ft according to p ( Ft|sgt , st ) 14: update subgoal prior distribution by posing p ( SGt = st ) = 0 15: end while The distribution p ( SG ) is initially determined by using Eq ( 1 ) to assign an algorithmic prior value to each state of the environment . Therefore , we sort in descending order the values of p ( SG ) and impose that P ( S G = s g o a l ) ∝ max ( p ( S G ) ) + Δ ( 7 ) where Δ > 0 is the maximum difference , in absolute value , between two consecutive priors arranged by the ordering . Note that the maximization of Eq ( 7 ) operates over the values of p ( SG ) calculated using Eq ( 1 ) , which are predefined during the inference . Afterwards , the whole prior distribution is normalized taking into account the modified value for P ( SG = sgoal ) . At time t = 0 , a subgoal sg0 is drawn from the a priori algorithmic probability distribution ( line 3 ) presented in the Section on “Algorithmic priors” and modified as in Eq ( 7 ) . Then , the instruction in line 4 searches for a policy πt such that it is possible to build a program representing a transition from st to sgt . This is achieved by drawing a policy from the following probability distribution: p ( Π t = π j | s t , s g t ) ∝ p ( s g t , s t | Π t = π j ) p ( π j ) . ( 8 ) Eq ( 8 ) expresses the probability of selecting a specific policy πj in function of the length of the program code that from the current state st , by means of πj , brings to the subgoal sgt , weighted with the prior of the policy πj It is possible to rewrite Eq ( 8 ) in an algorithmic form as: p ( Π t = π j | s t , s g t ) ∝ p ( s g t , s t | Π t = π j ) p ( π j ) ∝ 2 - | p s g t ( s t , π j ) | p ( π j ) ( 9 ) where the first factor of the right side product derives from the algorithmic joint probability shown in Eq ( 3 ) by considering the policy πj as fixed . Hence , in the resulting Eq ( 9 ) , the probability of a specific policy πj is proportional to its capability to generate a program that starts from the current state st and reaches the currently selected subgoal sgt . Note that Eq ( 9 ) is symmetric in the route as the transitions are symmetric—but the fact that we assign the final goal a high prior gives inference a directionality . According to Eq ( 9 ) , the probability of a policy π conditioned by a state s and a subgoal sg is the product between the likelihood that π generates a path from s to sg and the a priori probability p ( π ) . By assuming that p ( π ) is uniform , for any given pair of s and sg , there exists a subset of policies that have the same probability . In our simulations , drawing from the distribution defined in Eq ( 9 ) takes place by means of a Monte Carlo sampling method [70] . The candidate policies are sampled from the uniform distribution p ( Πt ) and then the policy corresponding to π* = arg maxj p ( πj|st , sgt ) is selected . This is the best policy ( i . e . , the one resulting in shortest paths ) amongst those that were sampled . Of note , various alternatives to sampling methods ( e . g . , heuristic techniques or tree search [71] ) can be adopted . The inference described so far starts from an initial ( clamped ) state st and returns a subgoal sgt and a policy πt able to reach it . ( This can be viewed as an Option-like plan built on-the-fly with the minimum number of actions involved into the transition from the initial to the subgoal state . ) Given the policy πt , and knowing the state st , the action at is determined in a straightforward manner ( line 5 ) . At this point—line 6—the node Ft checks the reaching of a goal state sgoal . If st = sgoal then ft = 2 and inference process stops , otherwise it proceeds until at least one among the termination criteria is fulfilled ( line 7 ) : either the node Ft evaluates to 2 or a maximum number Tmax of inferential cycles has been effected . During these ( from line 7 to line 15 ) and after that the next state st+1 is determined via p ( st+1|st , at ) Table 1 ( line 9 ) , the subgoal transition SGt → SGt+1[38] is established ( line 10 ) . In case that st ≠ sgt , the node Ft assumes a zero value and the subgoal sgt+1 is forced to be the same as time t . Contrarily , when ft = 1 , the current state st is equal to the current subgoal and a new one must be found . In order to guide subgoal determination towards the goal state sgoal , the inference “clamps” the current subgoal sgt , and assumes that ft+1 = 2 , namely it fictively considers the goal state as observed . Therefore , by the aforementioned considerations , the distribution p ( SGt+1 | ft+1 = 2 , sgt ) can be stated as: p ( s g t + 1 | f t + 1 = 2 , s g t ) ∝ p ( f t + 1 = 2 | s g t + 1 ) p ( s g t + 1 | s g t ) ≡ p ( s g o a l | s g t + 1 ) p ( s g t + 1 | s g t ) ( 10 ) The term p ( sgt+1 | sgt ) estimates the probability that the subgoal sgt+1 is chosen after sgt and the likelihood p ( ft+1 = 2 | sgt+1 ) corresponds to the conditional probability p ( sgoal | sgt+1 ) of the goal state sgoal with respect to the subgoal sgt+1 . The algorithmic expression of Eq ( 10 ) stems from conditioning all the programs returning the goal to produce sgt+1 as intermediate outcome; thus , it becomes: p ( s g t + 1 | f t + 1 = 2 , s g t ) ∝ p ( s g o a l | s g t + 1 ) p ( s g t + 1 | s g t ) ∝ 1 P ( s g t + 1 ) P ( s g t ) ∑ j 2 - | p s g o a l ( s g t + 1 , π j ) | ∑ j 2 - | p s g t + 1 ( s g t , π j ) | ( 11 ) Candidate subgoals sgt+1 are sampled by a Monte Carlo process [70] from the subgoal prior distribution p ( SGt ) . The subgoal at time t + 1 is calculated as sg* = arg maxk p ( sgk | f t+1 = 2 , sgt ) . Summing up , the SGt+1 subgoal state is determined by means of a posterior estimation in dependence on the values of st and sgt , according to the equation: p ( S G t + 1 | f t , s g t ) = δ s g t + 1 , s g t if f t = 0 p ( S G t + 1 | f t + 1 = 2 , s g t ) if f t = 1 δ s g t + 1 , s g o a l if f t = 2 ( 12 ) where δa , b is 1 when a = b while is 0 otherwise . Once the state for SGt+1 has been set , the inference proceeds from line 11 to line 13 by , according to this sequence and following the methods previously discussed , selecting a policy πt+1 , an action at+1 , and assessing ft+1 . Finally , in the instruction at line 14 , subgoal prior distribution is updated by ‘switching off’ the prior of the current state , i . e . , setting the probability p ( SGt = st ) to zero and normalizing the whole distribution . This update has the effect of an ‘on-line memory’ preventing the inference from selecting at the line 10 previously visited states as subgoals . In sum , the output of the iterative sampling procedure described in Algorithm 1 is a planning sequence Seq . Importantly , this procedure is conducted in parallel by multiple particles of a particle-filtering algorithm ( with resampling ) [70] . Each particle runs through all the problem space ( for a maximum number of steps Tmax ) and returns a sequence of subgoals and associated programs that solve the problem ( for the sake of simplicity , this ensemble of particles continues inferring paths until a percent Gres of them achieves the goal ) . Note that although the subgoaling procedure usually splits the problem into smaller sub-problems , it is also possible that a given particle finds a solution that does not require subgoals . In the next subsection we describe how the to-be-executed plan is selected based on a mechanism that accumulates the “votes” of all the particles that solve the problem . The inferential procedure illustrated in Algorithm 1 permits to probabilistically solve planning problems . Nevertheless , its nature is essentially stochastic because a sampling technique is used to draw both policies and subgoals . Consequently , given the same pair of goal and start states , the model will infer more than one possible path as a response . This creates a problem of plan selection , because ultimately the agent can only select and execute one single plan . We model this selection problem in terms of an accumulation-to-bound process , in which at every iteration each particle reaching the goal brings a “vote” for its candidate plan . Specifically , we use a variant of the widely adopted drift-diffusion model [72] that casts choice as a Bayesian sequential analysis over multiple alternatives [63 , 64 , 73] . In this approach , a set of different planning hypotheses is sequentially analyzed by gathering evidence for each of them during their execution . The sequential analysis is stopped when the results become significant according to a pre-defined termination criterion . The posteriors for the planning hypotheses are updated by using Bayes’ rule until reaching a threshold value . This method is implemented via a particle filtering algorithm with resampling reported in Algorithm 2 ( see [70] ) . At each resampling step r , for a maximum number of R , a set Qr = {q ( 1 ) , … , q ( K ) } of particles representing K distinct planning hypotheses of the internal model activation is generated ( line 7 ) and tracked forward until a time TR . By applying Algorithm 1 at the line 10 , a subset of H ≤ K particles , able to successfully infer a plan directed to the goal , is carried out and the related subgoal sequences Seqh = [sgt = 0 , … , sgt = Tr]h , for h = 1 , … , H , are evaluated by means of a score θh . Algorithm 2 Sequential Decision Making ( s0 , sgoal , K , Tmax , Θ , R , Gres ) Require: Starting state s0 , goal state sgoal , particle number K , maximum number of forward inferences Tmax , decision threshold Θ , number of resamplings R , particles-gone-to-goal threshold Gres . Ensure: Subgoal sequence probability distribution . 1: r = 0 2: initialize the set of subgoal sequences Σ ≡ {Seqh}h = 1 , … , H ≤ K = ∅ 3: initialize θh ( 0 ) 4: compute p0 ( SG ) by means of Eq ( 1 ) and Eq ( 7 ) 5: while ( θh ( r ) ≤ Θ , ∀h and r < R ) do 6: k = 0 7: create the sample set Qr = {q ( 1 ) , … , q ( K ) } 8: while ( k ≤ K and G ( Qr ) <Gres ) do 9: k = k+1 10: Seqh = Planning Inference ( s0 , sgoal , q ( k ) , pr ( SG ) , Tmax ) 11: Σ = Σ ∪ Seqh 12: end while 13: r = r + 1 14: update votes θh ( r ) for the sequences in Σ set by Eq ( 13 ) 15: update subgoal prior distribution pr ( SG ) by Eq ( 14 ) 16: end while Initially , when r = 0 , we assume that pr ( SG ) is computed through Eq ( 1 ) and Eq ( 7 ) and that , additionally , θh ( 0 ) = p ( Seqh ) where p ( Seqh ) is the prior distribution on the potential subgoal sequences . It is possible to assign a prior distribution for the sequences on the basis of the information extracted from the specific state space; on the other hand , this prior can be flattened when no additional information is present . This inferential process is executed until at least one of the inference terminating conditions is reached ( from line 8 to line 12 ) : either G ( Qr ) ( the percent of particles in Qr reaching the goal ) is greater than or equal to a given threshold Gres , or the inference reaches a ( fixed ) maximum number Tmax of iterations . Subsequently , the scores of the subgoal sequences Seqh are updated at the line 14 using the recursive formula: θ h ( r + 1 ) ∝ θ h ( r ) · p Q r | S e q h ( 13 ) where p Q r | S e q h = 1 K ∑ k K p q ( k ) | S e q h is the evidence of the current particle set Qr given the sequence Seqh , computed as the proportion of particles qk tracing the specific subgoal sequence Seqh . Therefore , the scores θh have the meaning of posteriors on Seqh accumulated in r steps . Each step r concludes by updating , in line 15 , the subgoal priors pr ( SG ) on the basis of the sampled sequences: p r + 1 S G = s g k ∝ p ( Q r | s g k ) · p r ( S G = s g k ) ( 14 ) where p ( Qr|sgk ) is the rate of the particles numbering sgk among the subgoals exploited . At the successive step r + 1 , a new set of hypotheses Qr+1 is resampled by adopting pr+1 ( SG ) as subgoal prior distribution , which increases the probability of selecting the more effective subgoal sequences at the next step . This procedure is iterated until one of the convergence criteria is met ( line 5 ) : ( a ) a subgoal sequence receives a total score greater than or equal to a predefined decision threshold Θ , or ( b ) the maximum number of iterations R is performed . In both cases , the sequence with the highest score is selected for execution . To verify if the proposed computational model can successfully reproduce human problem solving strategies , we tested it in three representative ToH tasks that are widely used to study how humans solve ( or fail to solve ) structured problems .
A first important finding that we address here is the fact that subjects that solve a ToH have been found to be very sensitive to the community structure of the problem [74] . As shown in Fig 1C , a ToH has a community structure composed of three clusters of nodes separated by so-called “bottlenecks” , viz . states S9 − S11 , S6 − S10 , and S23 − S24 . The bottlenecks are here defined topologically as narrow segments bridging between densely interconnected clusters of vertices; in other words , bottlenecks are the only way to pass from one cluster to another . For example , the bottleneck S9 − S11 is the only way to pass from the top cluster to the bottom-left cluster or vice versa . Previous research has identified the importance of community structures in carving problem spaces [75] and the ToH is no exception . In a series of empirical studies on human problem solving [74] , participants were asked to solve a ToH problem equivalent to navigating from S11 ( starting state ) to S13 ( goal state ) , see Fig 1C . It is possible to note that there are two shortest-path solutions to the problem that require the same ( minimal ) number of steps: the former is w1 = 〈S11 , S14 , S18 , S24 , S23 , S17 , S13〉 and the latter is w2 = 〈S11 , S9 , S8 , S7 , S6 , S10 , S13〉 . If the number of steps were the only determinant of behavior , participants should select the two paths with the same probability . However , there is an important difference between the two paths: the former requires traversing one bottleneck ( i . e . , one boundary between two communities , S23 − S24 ) while the latter requires traversing two bottlenecks ( S9 − S11 and S6 − S10 ) . Thus , if participants take the community structure into consideration when they plan and prefer remaining in the same cluster of nodes ( i . e . , not traversing bottlenecks ) when possible , they should prefer the former solution ( w1 ) with higher probability . Participants selected w1 in 72% of the cases , suggesting that they are sensitive to the community structure of the ToH and prefer remaining in the same community . The fact that there are two shortest-path solutions but one is consistently preferred indicates that the probability to select a path does not simply depend on its “physical” length ( i . e . , the number of steps ) . Instead , it might be also sensitive to the informational / community structure of the environment ( e . g . , the transition probabilities or the presence of community boundaries ) . To put it in another way , in problem solving , a path between two points ( start and end of the problem ) is not necessarily represented ( or calculated ) in a metrical space that only encodes their physical distance . Instead , it might be represented in a subtler “problem space” in which—say—the length of a path or the probability to select it depends on additional factors such as informational constraints , the cost of the inferential process required to solve the problem ( e . g . , find the path ) or the amount of information required to encode or recall it [43–45 , 74 , 75] . Our proposed model incorporates a new hypothesis on how structural / informational constraints are incorporated in the inferential process . In our probabilistic approach , the prior subgoal probability shown in Fig 1C permits to identify bottlenecks or boundaries between communities in terms of paths that have low probability to be traversed . As the subgoal ( prior ) probability of the bottleneck nodes is smaller than the other nodes , any path traversing bottleneck nodes will have lower probability in our inferential system , discouraging subjects from crossing boundaries or traversing clusters of nodes . This stems from a key aspect of the algorithm: states having lower prior probability are more rarely sampled as subgoals during the subgoal selection phase ( line 3 of Algorithm 1 ) . Accordingly , the inferential system tends to select paths that have higher probability ( or equivalently , correspond to shorter programs ) , which in turn are preferentially composed of high-probability states . Indeed , the probability of states enters into the evaluation of the policies that generate paths , see Eq 9 . Note also that the prior probability distribution is graded , and the probability of nodes decreases in proximity to bottlenecks . This implies that not only traversing bottlenecks , but also going towards bottlenecks is less probable in our inferential system . To illustrate the effects of this subgoal-induced community structure in quantitative terms , we compared our model with subjects’ performance as reported in [74] . Let’s consider again the problem of starting from S11 to reach the goal state S13 ( see Fig 1C ) . Considering the two possible shortest-path solutions , ( w1 = 〈S11 , S14 , S18 , S24 , S23 , S17 , S13〉 with only one community-traverse , and w2 = 〈S11 , S9 , S8 , S7 , S6 , S10 , S13〉 with two community-traverses ) , the first path is made up of states with higher subgoal prior probability . Our simulations show that even in the presence of these two shortest-path solutions to this problem , each involving the same number of steps , the winning ( i . e . , most voted ) strategy is w1: the one that selects the path traversing less bottlenecks or clusters . Fig 3 shows a dynamical competition between the two solutions in which particles “vote” for one of them ( see Section on ‘Methods’ ) . At each iteration ( resampling ) of the voting procedure the probability for w1 increases; note that it reaches the same proportion ( 72% ) as reported empirically in [74] after four iteration steps . Of course , the exact fit of the data is not extremely important here , nor is the specific combination of start and goal states , as our findings generalize to any other problem in the ToH . What our results show is that the proposed method reproduces the subjects’ sensitivity to the community structure of the ToH by only appealing to probabilistic computations and a principled approach to establish which subgoals are potentially useful . The three-cluster structure described above is not the only community structure of the ToH . Rather , the ToH has a 3-level nested community structure with ( triangular ) sub-clusters , see Fig 1C . Nested within the aforementioned ( level-3 ) clusters one can find three ( level-2 ) clusters ( e . g . , in the top level-3 cluster , these correspond to {S1 , S2 , S3} , {S5 , S8 , S9} , and {S4 , S6 , S7} ) . Furthermore , nested within each level-2 cluster one can find three level-1 clusters that correspond to individual nodes ( e . g . , in the top level-2 cluster , the nodes S1 , S2 , and S3 ) . Bottlenecks at different levels of depth correspond to these nested clusters: S2 − S3 is a level-1 bottleneck , S7 − S8 is a level-2 bottleneck , and S23 − S24 is a level-3 bottleneck . It has been reported that this nested structure affects human behavior [74] . Specifically , the costs for traversing a level-3 bottleneck ( S23 − S24 ) are higher than those for traversing a level-2 bottleneck ( S7 − S8 ) , which in turn are higher than those for traversing a level-1 bottleneck ( S2 − S3 ) —where the costs are implicitly measured as longer reaction times required to make a decision in the ToH . Once again , this difference only exists in a “problem space” and not in the standard “metric space” that only measures the number of steps to reach a goal location , because traversing a bottleneck ( independent of its level ) only requires one step . The sensitivity for the nested ToH structure can be explained within our framework if one considers that the prior probability of traversing S2 − S3 ( P ( S3|S2 ) ⋅ P ( S2 ) = 0 . 0464 ) is higher than traversing S7 − S8 ( P ( S8|S7 ) ⋅ P ( S7 ) = 0 . 0358 ) , which in turn is higher than traversing S23 − S24 ( P ( S24|S23 ) ⋅ P ( S23 ) = 0 . 0355 ) , see Fig 1C . These results thus extend those reported in the former Section and illustrate how the structure of the prior subgoal distribution nicely captures key characteristics of the “problem space” that humans use to solve problems . Empirical studies of how humans solve the Tower of Hanoi have identified a specific deficit in the failure of executing counterintuitive movements: moves that are apparently in opposition to the end goal-state ( e . g . , remove a disk from the target rod ) but that are necessary to achieve the goal efficiently . In these cases , two strategies have been identified that have opposite results: a “look-ahead” strategy that considers the long-run effects of the counterintuitive movements and their benefits for the overall problem solving , versus a “perceptual” strategy that only tries to decrease myopically the perceived ( apparent ) distance to the goal state ( e . g . , only increases the number of disks in the target rod ) [57 , 76] . This latter strategy disregards counterintuitive moves—which , by definition , require removing a disk from the target rod , thus apparently increasing the perceived distance from the goal state ) and can lead to suboptimal behavior . Here we characterize both strategies from a common probabilistic inference viewpoint , and analyze why the former ( look-ahead ) permits efficient problem solving while the latter ( perceptual ) determines a failure of executing counterintuitive movements . Let’s consider a sample problem consisting in going from a start ( S27 ) to a goal ( S20 ) configuration , see Fig 1 . In our implementation , the two aforementioned ( “look-ahead” vs . “perceptual” ) strategies only differ for the choice of prior distributions of subgoals p ( SG ) . The priors for the “look-ahead” strategy are shown in Fig 4A ) ; they are calculated using the same approach as used in the former two studies . The priors for the “perceptual” strategy are shown in Fig 4B . Here , in keeping with the problem solving literature , the probability of a movement is computed using a perceptual-based proximity criterion: the more disks the agent sees on the correct rod ( in the example of Fig 1 , the central rod ) as a consequence of the movement , the more probability it assigns to the movement ( see Methods for details ) . Once these priors are set , the two strategies use the same probabilistic inference methods introduced before—thus , they only differ for their choice or priors , not the inference . Because they use the same inference methods , the performance of the two strategies , which are based on considerations of plan optimality ( “look-ahead” ) or perceptual proximity to the goal ( “perceptual” ) can be directly compared . Furthermore , it is possible to run experiments that consider various ( weighted ) combinations of look-ahead and perceptual strategies . Combined strategies are created by allocating a percentage of the particles to each of the two strategies during the inference ( e . g . , 50% of the particles use the priors of the “look-ahead” strategy , and the remaining 50% use the priors of the “perceptual” strategy ) while also preventing any resampling during the inference . Fig 4C illustrates the simulation results for various strategies , which range from a “pure” perceptual strategy ( left ) , where 100% of the particles use the priors of the “perceptual” strategy , to a “pure” look-ahead strategy ( right ) , where 100% of the particles use the priors of the “perceptual” strategy , and all the intermediate cases ( e . g . , 50% of the particles use the priors of the “perceptual” strategy , and 50% of the particles use the priors of the “look-ahead” strategy ) . Parameters are reported in Table 1 . Our simulations show that the percent of particles able to find the shortest plan from state S27 to state S20 ( i . e . , a plan that only includes 7 moves ) varies as a function of how many particles of the particle filtering algorithm use the “look-ahead” or “perceptual” strategies . A pure “look-ahead” strategy is the most successful while the performance degrades quickly when increasingly more particles are allocated to the “perceptual” strategy . These results thus speak to an advantage of the “look-ahead” over the “perceptual” strategy , where their differences here are explained in terms of different prior distributions , not of different inferential mechanisms . The dynamics of the subgoal probability distributions p ( SG ) of the “look-ahead” strategy are shown in Fig 5 . The first panel shows the prior SG distribution once the start and goal states are set . The successive panels show how this distribution is updated during the inference , reflecting the fact that the particles are approaching the goal location . Effectively , already from the second panel all the high-probability subgoals lie in the best path from the start to the goal . Furthermore , it is evident that during the inference , the “target” candidate subgoals are increasingly closer to the goal location—and the goal location is the highest probability location ( only ) in the last panel . In other words , this algorithm implicitly creates a “moving target” or “gradient” of intermediate subgoals that permit splitting the problem into more manageable subproblems , carving the huge search space ( ∼ 1012 possible policies ) . This situation can be contrasted with the dynamics of the subgoal probability distributions p ( SG ) of the “perceptual” strategy , shown in Fig 6 , which lacks an equivalent gradient . Different from the “look-ahead” strategy , the “perceptual strategy” appears to be impaired by strong priors close to the final goal . It is this bias that produces a myopic strategy and prevents subjects from performing counterintuitive moves that would apparently move farther from the goal state . These differences between the two strategies have significant effects on their relative performance . The most frequent solution found by the particles using the look-ahead strategy requires 7 inferential steps while the most frequent solution of the perceptual strategy requires 9 inferential steps ( to understand how the most frequent or “most voted” solution is computed , see the Method section ) . The 7-steps solution found with higher probability by the look-ahead strategy is a unique , optimal plan passing through the states: S27 , S26 , S25 , S24 , S23 , S22 , S21 , S20 ( which , for illustrative purposes , we also show in the usual Tower of Hanoi format in Fig 7 ) . Note that as shown in Fig 4D most of the particles that find the aforementioned 7-step optimal plan use 2 or 3 subgoals , which highlight an advantage of subgoal-based strategies over strategies that try to solve a problem from start to end without splitting the problem space . To validate our approach , we used our model to reproduce the results of an empirical study that investigated how adult patients with lesions of the prefrontal lobe and a control group solve Tower of Hanoi puzzles of increasing difficulty [77] , see also [67] . The study revealed specific deficits in lesioned patients in resolving a goal-subgoal conflict , in particular in the case of counterintuitive movements requiring to move away ( in a perceptual sense ) from the goal state . Here we simulated three ToH problems analogous to those reported in [77] , which exemplify an easy , an intermediate , and a challenging situation . All the three problems ( P1–P3 ) had the same goal state S20 . In the easiest problem P1 , the initial state was S13 . This problem has a simple solution requiring only 3 moves . In the intermediate problem P2 , the starting state is S27 . The shortest solution of this problem requires 7 moves ( the maximum number in our Hanoi Tower ) and crosses a bottleneck between two community structures . In the hardest problem P3 , the starting state is S9 . Like the intermediate problem , the shortest solution to this problem requires 7 moves and crosses one bottleneck ( S6 − S10 ) . However , this path is harder to find for the presence of more counterintuitive moves , which hinder particularly the perceptual strategy . For example , an agent following the perceptual strategy will tend to do a transition from S9 to S11 , not S8 , because S11 is perceptually closer to the goal state S20 . This tendency can prevent the agent from finding the optimal path , which passes through S8 . In keeping with our previous arguments , we associated the behavior of lesioned patients vs . control group to a different choice of strategies , the former more “perceptual” and the latter more “look-ahead” . Importantly , we do not consider these two strategies to be computationally different ( as commonly assumed in the ToH literature ) ; rather , we model both using our inferential method , but using two different prior distributions for subgoals ( perceptual priors vs . algorithmic priors ) . Accordingly , we associate the behavior of lesioned patients vs . control group to two models . In the former model , corresponding to lesioned patients , 85% of the particles are initialized according to the perceptual prior and 15% are initialized according to the algorithmic prior . in the latter model , corresponding to the control group , 85% of the particles are initialized according to the algorithmic prior and 15% are initialized according to the perceptual prior . Fig 8A shows the results of the experiment ( 25 executions of each model ) . As in the human experiment , we measure success rate as the percentage of participants ( or executions of the model ) that found the shortest path solution to the ToH problem . ( Note that in the human experiments , participants solve the experiment only once . ) In P1 , the success rate is high for both groups , but the variability of the solutions is higher for lesioned patients . In P2 and P3 , the success rate of both groups decreases , but—crucially—it does so more steeply for the simulated lesioned patients , reflecting their difficulties in facing challenging problems that include counterintuitive moves . Our simulations replicate the pattern of results reported in the empirical study [77] , and in particular the steeper decrease in the performance of lesioned patients when the problems require executing counterintuitive movements . This pattern of results cannot be explained by an algorithm that simply introduces noise in the best strategy , but results from different qualitative strategies , with the perceptual strategy that performs quite well in simple cases but not in the more complex ones that include counterintuitive movements . The slightly lower ( overall ) performance of real participants compared to our simulations might result from minor differences in the set up . Indeed , the three problems P1–P3 reported here correspond to the three problems P5 , P1 and P8 in [67 , 77] , after a mapping between the ToH problem space used in our simulations ( with three disks and three pegs ) and the slightly more challenging one used in the empirical report ( with five disks and three pegs ) . It is important to note that despite the minor differences in the set up , the structure of the problem is the same ( e . g . the number of bottlenecks and counterintuitive moves ) and the overall pattern of results of the simulations coherent with the real data . Besides replicating the pattern of results in [77] , our simulations can provide a more fine-grained analysis of the behavior of the two groups . The two matrices of Fig 8B and 8C show the number of solutions to problem P3 ( x-axis ) and relative number of “votes” ( grey scale ) found by the simulated lesioned patients and control group , respectively . It is possible to appreciate that the simulated control group is able to find many more solutions than the simulated lesioned patients . This result confirms that the look-ahead strategy is more flexible and permits agents to find a broad spectrum of solutions , rather than narrowing the problem space . This increased flexibility is especially advantageous in challenging problems like P3 and might explain the large gap in the performance between the two models , see Fig 8A . Furthermore , it is worth noting that the most voted solutions of the control group receive more votes than the most voted solutions of the lesioned patients ( compare the grayscale of Fig 8B and 8C ) , which in turn implies a faster convergence of the algorithm in the former case .
We have presented a formal approach to human-level problem solving , here exemplified in the Tower of Hanoi ( ToH ) task . We use probabilistic inference methods that are increasingly adopted to study multiple cognitive domains , such as perception , action and learning [17 , 18] , supporting the idea that the computations underlying problem solving might share common principles with them . Specifically , we leverage on the planning-as-inference framework ( PAI ) and extend it to address problem solving by introducing a crucial additional mechanism: subgoaling . Our emphasis on subgoaling is in keeping with their recognized importance of subgoaling in human problem solving and cognitive architectures [1 , 8–13 , 57] and with a vast computational literature showing that subgoals can carve the problem space and reduce the computational complexity of problems [14–16 , 47 , 66 , 78] . Our results illustrate that a subgoaling-based probabilistic inference approach can explain key aspects of human problem solving . Our first two studies focused on what structure or problem space humans use . Convergent findings indicate that the time to execute a ToH puzzle is proportional to the complexity of the problem , not to the number of steps in a graph . The complexity of the problem can be characterized as a distance in problem space between the start and goal configurations , which does not only consider the number of ( physical ) steps required to solve a problem , but also computational requirements ( e . g . , the complexity of solutions and associated computational costs ) , which in turn are influenced by “community structure” of the problem [74 , 75] . In the proposed computational approach , an important constituent of the problem space is prior subgoal distribution p ( SG ) —or an a-priori probabilistic estimate of the likelihood of traversing a given set of states . The first two studies thus show that the prior subgoal distribution measure can explain two typical ( but otherwise puzzling ) idiosyncrasies of human problem solving strategies , and their sensitivity to the ( community ) structure of the problem at hand . Our third and fourth studies show that a specific and well-documented deficit in human problem solving—the failure to execute counterintuitive movements—can be explained in terms of a mis-identification and mis-use of ( good ) subgoals within our probabilistic inference scheme . One interesting aspect of our proposal is that it permits to describe two competing strategies for solving the ToH , look-ahead vs . perceptual [57] , within a homogeneous probabilistic inference method , without appealing to two segregated mechanisms . In this perspective , when the perceptual strategy is used , the subgoal distribution quickly collapses into a narrowly-focused problem representation , in which the final goal dominates the inference , preventing subjects to carve the problem in useful ways . In this perspective , the strategy a person uses during problem solving ( and her errors ) might be predicted by looking at her subgoal distribution prior and during the inference . This is a novel empirical prediction that can be potentially tested in the ToH or related puzzles . It is worth noting that the disadvantages of the perceptual strategy might be partially compensated by the fact that calculating subgoal distributions using a simple “perceptual distance” between states could be less cognitively demanding than updating them according to the look-ahead strategy . In this perspective , the choice of a more accurate but also more cognitive demanding vs . a simpler but inflexible strategy might obey to computational trade-offs [19 , 40 , 79–81] . If this hypothesis is correct , introducing a cognitive load would shift the balance towards the latter ( perceptual ) strategy [82 , 83] . This prediction remains to be tested in future research . To summarize , we have shown that problem solving requires the ability to carve the problem space in useful ways , which do not only ( or not necessarily ) reflect a simple physical distance , but a distance in a subtler “problem space” . In this perspective , we have shown the advantage of representing possible subgoals in terms of algorithmic priors rather than in terms of a mere perceptual distance from the goal state . The way we calculate algorithmic priors P ( SG ) shares some similarities , but also differences , with alternative ways that have been proposed in the literature that are based on graph theory [74 , 75 , 84] and information theory [43–45 , 85 , 86] . In our approach , the algorithmic prior of a state considers , first , the number of policies that generate programs terminating in the state ( the more the policies , the higher the probability ) and second , the length of these programs ( the shortest the programs , the higher the probability ) —thus reflecting a prior preference for traversing every state by using the best ( shorter ) program . In the ToH maze , this second aspect of the algorithm tends to assign higher probability to states that are close to ( but are not ) vertexes , because the programs that start from vertexes are shorter in these states compared to all the other states , including bottlenecks ( mazes with different topologies will have different p ( SG ) distributions , of course ) . This implies that bottlenecks do not have the highest prior probability . This might seem in contrast with a graph-theoretic perspective , where bottlenecks are usually identified as salient structural aspects of the problem [74 , 75 , 84] . However , two points are in order . First , graph-theoretic algorithms do not assign a probability distribution over all states , but only identify bottlenecks , thus it is hard to directly compare the two methods . Second , in our approach p ( SG ) reflects algorithmic probability measures and the prior propensity to visit or traverse a state rather than its perceptual salience or connectedness; and we have shown that this choice of priors permits to model accurately human behavior , especially the preference for paths that include lesser ( or lower-rank ) bottlenecks . It is possible that graph-theoretic measures and our approach based on algorithmic probability reflect distinct ( not necessarily divergent ) structural aspects of the problem , the former more revelatory of topological aspects of the maze and oriented towards ( optimal ) task decomposition [74] , and the latter more related to informational constraints of the problem to be solved—to which humans seem to be particularly sensitive . Supporting this idea is the fact that our approach is largely convergent with other methods that identify subgoals using info-theoretic measures such as the information bottleneck method [85 , 86] and relevant goal information ( RGI ) [43–45] . The RGI method uses the mutual information between goal and action to assign states an information value; like our approach , this method identifies transition points , and assigns bottlenecks low ( not high ) information value , and high information value to bottleneck neighbors . While using different information measures , the RGI method is largely convergent with our approach as it reflects informational constraints of the problem ( e . g . , the amount of information an agent needs to maintain about its goal and the points where she needs to change subgoals ) that—we argue—are important determinants of human problem solving in challenging tasks such as the ToH . The methods we use are also related to Hierarchical Reinforcement Learning ( HRL ) [48] . In particular , the transition p ( π|s , sg ) is related to the concept of an “Option” in HRL [48] but it is expressed probabilistically . In HRL , learned Options influence policy selection and guide the agent transitions for an extended period of time , usually up to a predefined subgoal ( e . g . , an Option might correspond to “move until the next door” ) . Rather , here there are no predefined or “cached” subgoal-and-policy pairs . Policies are sampled at each step of the inference , while subgoals are sampled whenever the previously selected subgoal has been reached . Thus , in a sense , this system forms an Option-like structure on-the-fly that guides the agent’s transitions up to the next predefined subgoal . In principle , the ( best ) Option-like structures ( e . g . , p ( π|s , sg ) ) formed during the inference might be “cached” to facilitate future inferences; this is something we plan to explore in future studies .
We presented a novel computational theory of human problem solving that is based on probabilistic inference augmented with a subgoaling mechanism . Probabilistic inference methods are increasingly used to explain a variety of cognitive , perceptual and motor tasks , including goal-directed decisions and planning [17 , 23 , 30 , 34 , 37 , 38 , 87] . Here we show that probabilistic inference , when enhanced with a subgoaling mechanism , can explain various aspects of human problem solving , too , including its idiosyncrasies and deficits , such as the human sensitivity to the structure of the problem space , and patient deficits in handling counterintuitive moves and goal-subgoal conflicts . We focused on the well-studied Tower of Hanoi ( ToH ) task , which has been modeled using several computational frameworks such as the cognitive architecture ACT-R [88] and various symbolic [81 , 89] and subsymbolic systems [90] , none of which however use the principles of probabilistic inference proposed here . This computational analysis suggests that human problem solving does not necessarily need to be considered a special ( ized ) domain or module of cognition , but could use the same probabilistic computations that are widely studied in other fields of cognitive science . Along similar lines , it is not necessary to assume that suboptimal strategies such as the perceptual strategy are mechanistically different from the optimal solution or heuristics . Our study shows that they can be explained within the probabilistic framework introduced here under the assumption of a different problem ( prior ) representation . This computational analysis underlies the importance of subgoaling in problem solving , too , showing that subgoals—as expressed for example in the prior subgoal distribution p ( SG ) —permit to carve the problem space in useful ways . Importantly , subgoals define a metric for the problem that is sensitive to the probability of transitions between states rather than to the mere count of the number of states from start to goal . Humans are sensitive to key aspects of this metric , such as its community structure [74] . Furthermore , subgoals can be used as “waypoints” that permit finding parsimonious solutions: the algorithm splits the problem into smaller subproblems , each requiring less information to be encoded compared to a solution from start to end ( this adds on to the fact that the inferential method tends to select programs having higher algorithmic probability and thus having a shorter code-length [65 , 66 , 68 , 69] ) . From a cognitive perspective , the overall divide et impera strategy can be less taxing for an agent , because she only needs to “remember” a part of the solution ( e . g . , the path to the next subgoal ) at every moment in time . Reaching a subgoal ( as signaled by the node F in our model ) marks a transition to a next subgoal , and implies that the agent can “forget” past information [43–45] . Furthermore , finding waypoints permits to learn and store subroutines ( e . g . Options in HRL ) that—in principle—can be reused in future planning problems , thus saving resources [91–93] . Finally , the failure to use subgoal information in appropriate ways can explain specific problem solving deficits such as the inability to execute counterintuitive movements of prefrontal patients [57] , without appealing to separate cognitive mechanisms for look-ahead vs . perceptual strategies . | How humans solve challenging problems such as the Tower of Hanoi ( ToH ) or related puzzles is still largely unknown . Here we advance a computational model that uses the same probabilistic inference methods as those that are increasingly popular in the study of perception and action systems , thus making the point that problem solving does not need to be a specialized module or domain of cognition , but it can use the same computations underlying sensorimotor behavior . Crucially , we augment the probabilistic inference methods with subgoaling mechanisms that essentially permit to split the problem space into more manageable subparts , which are easier to solve . We show that our computational model can correctly reproduce important characteristics ( and pitfalls ) of human problem solving , including the sensitivity to the “community structure” of the ToH and the difficulty of executing so-called “counterintuitive” movements that require to ( temporarily ) move away from the final goal to successively achieve it . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Conclusions"
] | [
"learning",
"decision",
"making",
"applied",
"mathematics",
"social",
"sciences",
"problem",
"solving",
"neuroscience",
"learning",
"and",
"memory",
"simulation",
"and",
"modeling",
"algorithms",
"cognitive",
"psychology",
"mathematics",
"probability",
"distribution",
"co... | 2016 | Problem Solving as Probabilistic Inference with Subgoaling: Explaining Human Successes and Pitfalls in the Tower of Hanoi |
Epileptic seizure dynamics span multiple scales in space and time . Understanding seizure mechanisms requires identifying the relations between seizure components within and across these scales , together with the analysis of their dynamical repertoire . Mathematical models have been developed to reproduce seizure dynamics across scales ranging from the single neuron to the neural population . In this study , we develop a network model of spiking neurons and systematically investigate the conditions , under which the network displays the emergent dynamic behaviors known from the Epileptor , which is a well-investigated abstract model of epileptic neural activity . This approach allows us to study the biophysical parameters and variables leading to epileptiform discharges at cellular and network levels . Our network model is composed of two neuronal populations , characterized by fast excitatory bursting neurons and regular spiking inhibitory neurons , embedded in a common extracellular environment represented by a slow variable . By systematically analyzing the parameter landscape offered by the simulation framework , we reproduce typical sequences of neural activity observed during status epilepticus . We find that exogenous fluctuations from extracellular environment and electro-tonic couplings play a major role in the progression of the seizure , which supports previous studies and further validates our model . We also investigate the influence of chemical synaptic coupling in the generation of spontaneous seizure-like events . Our results argue towards a temporal shift of typical spike waves with fast discharges as synaptic strengths are varied . We demonstrate that spike waves , including interictal spikes , are generated primarily by inhibitory neurons , whereas fast discharges during the wave part are due to excitatory neurons . Simulated traces are compared with in vivo experimental data from rodents at different stages of the disorder . We draw the conclusion that slow variations of global excitability , due to exogenous fluctuations from extracellular environment , and gap junction communication push the system into paroxysmal regimes . We discuss potential mechanisms underlying such machinery and the relevance of our approach , supporting previous detailed modeling studies and reflecting on the limitations of our methodology .
Epilepsy is characterized by seizures , a paroxysmal behavior that results from abnormal , excessive or hypersynchronous neuronal activity in the brain [1] , with a various set of symptomatic outcomes depending on brain regions involved in its generation and propagation processes . Clinically , epilepsy affects 1% of the population , from whom 30% are drug-resistant . Physiological investigations of neural tissue in the context of human Temporal Lobe Epilepsy and experimental models revealed neuronal loss in the hippocampus , rewiring of excitatory and inhibitory pathways [2] , in keeping with the hypothesis on unbalanced excitation/inhibition ratio observed in epilepsy [3 , 4] . Understanding seizure mechanisms from micro to macro scales is necessary to provide clinicians and basic scientists with a reliable theoretical basis to develop new therapeutic approaches . Computational modeling reproducing brain activity is a genuine approach to investigate such multi-scale paradigms . Neural network models in the context of epilepsy typically use multi-compartment Hodgkin-Huxley type neurons with a collection of ion-channels dynamics and multiple excitatory and inhibitory synaptic combinations . We will here refer to them as biophysically-realistic , see for example [5 , 6] . Reduced population models ( so-called neural masses or mean field models ) absorb a significant amount of biophysical details in constant parameter values and are referred to as large-scale or macroscopic [4 , 7–9] , see for a review [10] . A third type of modeling scheme consists in approaching the dynamics of seizures in an abstract manner , and describing them in terms of generic dynamic features [11] . The advantage of this approach is its generality , allowing the identification of invariant seizure classes based on basic dynamical properties . The drawback lies in the difficulty to find biophysical correlates to the state variables used in such an approach . Certain elementary features such as dynamics evolving on different time scales guides the identification of the biophysical correlates . For example , recent emphasis in seizure modeling is directed towards the role of extracellular environmental fluctuations , which evolve on a significantly slower time scale than neuronal discharges . By incorporating slow extracellular potassium or oxygen levels as key parameters , state-of-art studies displayed transitions between pathological brain states observed during paroxysmal activity [12–14] . Such approaches combine dynamical systems theory and large-scale neural network computations to propose key insights into seizure mechanisms . However , extracellular potassium homeostasis provides only a partial answer . Many different biophysical factors can lead to seizure genesis [15 , 16] , in keeping with the concept that different parameter sets can produce the same type of activity at the network level [17] . Introducing all those parameters in a detailed model poses a computational and theoretical challenge , but may be only useful if the arising network behavior can be characterized dynamically . In the present study , we make the link of the seizure state dynamics across neuron and population levels explicit . Fig 1 illustrates the line of thought deriving the here presented intermediate system architecture ( panel B ) from the phenomenological model ( panel A ) ; future work could comprise spatially structured networks of neurons ( panel C ) . We specifically develop a network of inhibitory spiking and excitatory bursting neurons , driven by a slow environmental variable , reproducing characteristic features of the temporal evolution of human and experimental seizures . Inspired by the phenomenological model of spontaneous seizure generation from [11] , the so-called Epileptor , and mean field dimension-reduction techniques [18] , we derive the former equations as introduced in [11] from more biophysically-inspired representations of the system , including single neuron dynamics , linear and non-linear synaptic interactions portraying gap junctions and chemical synapses , respectively . Meanwhile adding this new level of complexity , we keep track of the emergent network behavior exhibited in the abstract model by systematically exploring dynamical changes induced by parameter sweeps in our new system . Our slow environmental variable is more abstract than the biophysically explicit extracellular milieu and oxygenation as considered in [12–14] and involves a wider set of physiological factors such as intra/extracellular pH ratio , oxygen availability , extracellular potassium , calcium , and chloride concentrations , for example . These parameters have been studied experimentally and demonstrated to be part of the influencing agents leading to seizures [3] . Introducing these variables into existing detailed biophysical models has been subject to great consideration for the last decade [19–22] . The abstract integration of such local mechanisms into a slow environmental variable , as presented here , aims at describing the slow sub-system from a more conceptual perspective . Following this direction , large scale simulations of neural systems become possible at reasonable computational cost , but with sufficient accuracy to treat complex dynamical mechanisms , such as multi-clustered synchronization or entrainment , while implying biologically realistic synapses and firing patterns . We systematically identify synchronization regimes and reproduce potential routes through status epilepticus ( by definition , a seizure lasting more than 20 minutes ) in our parameter space , providing interpretation of its underlying neuronal mechanisms and further validating our model . We accredit the different parametric regimes experimentally against rodent data recorded in vivo and demonstrate that the different routes in parameter space are consistent with the theoretically predicted topology . Then , exploring a regime that generates spontaneous seizures from background activity , we infer new insights from the role of excitatory and inhibitory synapses , as well as the extracellular environment .
The Epileptor [11] is a five-dimensional model and comprises three different time scales accounting for various electrographic patterns: On the fastest time scale , two state variables ( ensemble 1 ) exhibit bistable dynamics between oscillatory activity modeling fast discharges and a stable node representing interictal activity . On the intermediate time scale , two state variables ( ensemble 2 ) model the spike and wave events ( SWE ) and form the second neuronal ensemble . On the slowest time scale ( order of tens of seconds ) , the evolution of a very slow permittivity variable guides the neural population through the seizures including seizure onset and offset . The first ensemble is linearly inhibited by the second ensemble in order for fast discharges to occur only during the wave part of the SWE , the second ensemble is excited by the first ensemble through a low-pass filter coupling in order to generate SWE and interictal spikes . Both ensembles are coupled through the permittivity variable . A separatrix divides the state space of the first ensemble between ictal and interictal states and acts as a barrier . As the permittivity variable evolves over time , seizure onset occurs through a saddle-node bifurcation showing a direct current ( DC ) shift at the transition between the interictal and the ictal state . Seizure offset occurs through a homoclinic bifurcation showing the logarithmic scaling of interspike intervals when approaching seizure offset . The time course of the local field potential is related to the total activity of both ensembles . A more detailed description of the model can be found in Jirsa et al [11] , with an extended analysis of its embedded dynamics in [23] . We model the dynamics of brain activity across a set of two populations P1 ( excitatory ) and P2 ( inhibitory ) of N neurons each , x1 , j and x2 , j , respectively ( j∈[1 , N] ) . The dynamics of the population Pi is determined by x¯i=1N∑j=1Nxi , j as described by the mean field approximation used for neural masses [24] , and the dynamics of membrane potentials xi ( i∈[1 , 2] ) is defined by Hindmarsh-Rose ( Eq ( 1 . 1 ) , ( 1 . 2 ) and ( 1 . 3 ) ) and Morris-Lecar ( Eq ( 2 . 1 ) , ( 2 . 2 ) , ( 2 . 3 ) , ( 2 . 4 ) and ( 2 . 5 ) ) neuronal models for P1 and P2 , respectively , as follows . For sake of clarity , the jth index is omitted in the notation . Our choice of the Hindmarsh-Rose model is motivated by the fact that its phase flow is isomorphic to the phase flow of the first Epileptor ensemble [11] . This neuron model is a square-wave burster governed by saddle-node and homoclinic bifurcations , and so is the first Epileptor ensemble [25] . Similarly , the Morris-Lecar model , via its saddle-node-on-invariant-circle ( SNIC ) bifurcation , captures the same excitable properties as the second Epileptor ensemble . Population 1 neuron ( Hindmarsh-Rose ) : Population 2 neuron ( Morris-Lecar ) : CMV˙=I2−gL ( V−EL ) −gKn ( V−EK ) −gCam∞ ( V ) ( V−ECa ) +σ2CE ( x¯2−x2 ) +Isyn ( x¯1 , x2 ) +Isyn ( x¯2 , x2 ) −σ2×0 . 3 ( z¯−3 ) +σ2W ( t ) ( 2 . 1 ) n˙=ϕ ( n∞ ( V ) −n ) /τn ( V ) ( 2 . 2 ) with m∞ ( V ) =12[1+tanh ( ( V−V1 ) /V2 ) ] ( 2 . 3 ) τn ( V ) =1/cosh ( ( V−V3 ) /2V4 ) ( 2 . 4 ) n∞ ( V ) =12[1+tanh ( ( V−V3 ) /V4 ) ] ( 2 . 5 ) and Coupling term CE and function Isyn are detailed in a next section . σi is a scaling ratio between the two pools of neurons in order to have similar membrane potential amplitudes across different neuron types ( i∈[1 , 2] ) . Parameters are the same as in previously published studies [5 , 24 , 25] , unless otherwise mentioned . They are in part enumerated in Table 1 together with their biophysical interpretation , when applicable . W ( t ) is white Gaussian noise of uniformly distributed values in the interval [-Wmax , Wmax] where Wmax ranges from 2 . 5 to 20 mV as provided in Table 1 , 2 . 5 mV being used for lowest noise simulations and 20 mV for highest noise simulated traces . Here , all simulations were performed with bounding value of Wmax at 6 mV unless stated otherwise . The other parameters are constant and their ranges are included in Table 1 . All values are set at the initialization of the simulation and remain constant for the time of the run . I1 , I2 are baseline input currents of neurons for populations 1 and 2 , respectively . x0 captures the equilibrium point of the permittivity z ( Eq 1 . 3 ) and has previously been referred to as degree of epileptogenicity [26] in the context of the Epileptor . As mentioned above , it corresponds to the mutual effect of a set of factors influencing neural excitability including ATP availability , oxygenation , extracellular potassium concentration etc . We use symmetric linear difference coupling across membrane voltage equations on fast time scales to model gap junctions , also referred to as electrical synapses . This type of communication , being only electrically conductive through the cell's membrane , is considered nearly instantaneous so its transmission delay is negligible . It is described in the equations by the difference CE ( x¯i−xi ) where x¯i refers to the mean activity of the neural population i , and CE to the coupling strength . Note that when all neurons from a population are synchronized this difference is zero . Only neurons belonging to the same population are connected to each other via such linear coupling as it is observed in neural tissue that gap junctions usually connect neurons from the same class [27] . Chemically mediated synaptic transmissions follow more complex dynamics induced by intermediate biological mechanisms such as pre-synaptic neurotransmitter release , post-synaptic receptor binding , G-protein activation and so on . Models of such dynamics are typically non-linear and invoke physiologically most relevant parameters , including transmitters’ concentration and release time , conductance strength , or re-uptake time . We used the model described in [5 , 28] , as it provides a sufficiently accurate level of biological description for our study . The equations are: Isyn ( xi , xj ) =−GSi , ju ( xj−E ) ( 3 . 1 ) u˙=αT ( 1−u ) −βu ( 3 . 2 ) T=Tmax1+e− ( xi−Vt ) /Kp ( 3 . 3 ) where Isyn is the post-synaptic current , xi and xj are the pre- and post-synaptic neuron activity , respectively . u is an auxiliary variable for the computation of the post-synaptic current , E is the reversal potential , Gsi , j the conductance strength of synapses from neurons of population i to neurons from population j , with ( i , j ) ∈ [1 , 2] . α , β the forward and backward binding rate constants with transmitter concentration T in the synaptic cleft , of which the maximum is set by the constant Tmax . Kp gives the steepness and Vt sets the value at which the function is half-activated . We note that dendritic spatial summation corresponds to the process , in which the input xi is averaged over the whole population when sent to the synapse’ function Isyn . EEG time series of epileptic seizures display a large diversity of temporal characteristics such as interictal spikes , ripples , tonic/clonic discharges etc . It is therefore non-trivial to define a precise measure that delineates these features with sufficient accuracy . To analyze the output of our neuronal populations , we consider an index of synchronization between oscillators: the Kuramoto Order Parameter ( KOP ) [29] . To visualize the dynamics of the phases of each neuronal oscillator , it is convenient to imagine a swarm of points tracing out the unit circle in the complex plane when action potentials occur . The complex order parameter [30] is the norm of the sum of all vectors between the origin of the unit circle and the points around the circle . It is defined as follow: The norm r of this macroscopic quantity is near zero when action potentials are uniformly distributed over time , and increases as firings get synchronized . An animated version of these motions is presented in S1 Movie . It can be interpreted as the collective rhythm produced by the whole population . Although it is a convenient method to quantify synchrony , this measure is applicable only when oscillations are present . Neurons at rest ( i . e . not oscillating ) have been considered separately and labeled explicitly at rest in the analysis . Special care must be taken while interpreting our synchronization measure: we do not consider the synchronization between P1 and P2 but rather the synchronization within P1 ( KOP1 ) and within P2 ( KOP2 ) . The overall KOP is calculated as the sum of KOP1 and KOP2 . The code is written in Python with an object-oriented architecture , of which an online version is made available together with its documentation at the Github repository ( https://sebnaze@bitbucket . org/sebnaze/epilepton . git ) . Simulations were performed on a parallel computing cluster using the Euler-Maruyama integration scheme [31] with step size dt = 0 . 05 .
To build a system dynamically isomorph to the Epileptor model [11] , but with a link to biophysical properties , we constructed a network of Hindmarsh-Rose bursting and Morris-Lecar spiking model neurons . Fig 2 illustrates the dynamical similarities between the neuron models and the original Epileptor ensembles , and provides a comparison of their phase space topologies for various levels of excitation . In the Epileptor , the first ensemble shows a saddle-node bifurcation at seizure onset and a homoclinic bifurcation at seizure offset . The Hindmarsh-Rose model is a square-wave burster also governed by saddle-node and homoclinic bifurcations [25] and thus isomorph regarding its phase flow with the first Epileptor ensemble [11] ( Fig 2 , left column ) . The spike-wave discharge is modeled in the second Epileptor ensemble by a saddle-node on invariant cycle ( SNIC ) bifurcation , which is also present in the Morris-Lecar model [25] ( Fig 2 , right column ) . When the excitatory and inhibitory spiking neurons are electrically coupled within the populations via gap junctions , then synchronization occurs . By construction , the synchronization manifold is identical to the uncoupled neuron models and thus has the same bifurcations and phase flow topology as the Epileptor ensembles . This is valid precisely for full synchronization and approximately for partial synchronization as a function of the coupling strength . Due to the dynamic isomorphism of single neuron models and the Epileptor ensembles , when full synchronization is approached the mean field of the populations expresses the full dynamic range of behaviors known from the Epileptor model . This will be shown in the next section , in which a parameter space analysis maps out the synchronization behaviors to large detail . Status epilepticus ( SE ) , i . e . uninterrupted seizure lasting at least 20 min , is a traumatic experience that can transform a non-epileptic brain into a brain displaying spontaneous seizures [32 , 33] . SE can be induced in rodents by injecting convulsant agents , such as kainic acid or pilocarpine [34 , 35] . It develops through 3 main phases called impending , established and subtle SE ( Fig 3 , timeseries II , III and IV , respectively ) , each of which being part of a continuum of electrophysiological fingerprints [36] . In the following , we retrace the route that neuronal networks follow during SE through a selected set of physiological parameters i . e . the extracellular excitability x0 , inter- and intra-population synaptic coupling strengths Gsi , j and Gsi , i ( glutamatergic and GABAergic for pyramidal cell and interneuron populations , respectively ) and gap junction coupling strength CE ( Fig 4 ) . We identify regimes according to the synchronization ratio between neurons within populations ( color code ) , and map it to the characteristic phases observed experimentally ( Fig 3 and Fig 4 , roman numbers ) . Since a single metric has not yet been identified to discriminate appropriately the different electrophysiological regimes observed during SE , the color code does not separate the SE states but rather offers a coarse map for orientation amongst the observed dynamical regimes . There are no clear demarcation lines between the dynamic regimes , since the transitions are more gradual than discrete . After SE , animals experience a latent period during which complex network reorganizations take place . During such period , although neuronal networks exhibit interictal-like activity [38] , there are no spontaneous seizures . The latter occur during the chronic phase , a few days or weeks after SE . They are difficult to predict; the brain appears to operate “normally” before an abrupt change happens , characterized by 2 to 10-fold larger amplitude oscillations , which is the seizure . Our model reproduces the most important features of such transitions i . e . an abrupt fast firing discharge pattern at seizure onset , and a decrease of spike-wave frequency towards the end of seizure . We predict interictal spikes and spike-wave discharges are generated from synchronized activity of inhibitory neurons , and are affected by synaptic coupling strengths within and between the two populations of neurons . Fig 6 displays a simulation of about a minute of activity in which a seizure takes place , together with its experimental counterpart . The model produces the different states of seizure evolution without any change of parameters; the states include pre-ictal population spikes , abrupt transitions to tonic firing , and seizure offset . Hysteresis effects have been predicted in the Epileptor [11] and are preserved in the coupled neuronal population dynamics relayed by the slow permittivity variable . As permittivity traces out its trajectory , seizure onset and offset occur at different values of permittivity and the two different neuronal spiking patterns of the populations may co-exist for the same permittivity value . These behaviors are characteristic for hysteresis . Synaptic coupling is also supposed to play a central role in seizure genesis . However , extracellular Ca2+ concentration nearly drops to zero during seizures , including in primates [39] . In the absence of extracellular Ca2+ seizures can occur in neuronal networks [40] , following the general rules of seizure dynamics [11] . When glutamatergic and GABAergic synaptic couplings are removed in the model ( GSi , j = 0 in Eq 3 . 1 ) , we still observe seizure-like events but the temporal features of the signal are slightly different ( Fig 7 ) , although hysteresis is maintained . In such conditions , inhibitory populations’ spikes and pyramidal fast discharges influence each other’s excitation through the extracellular environment , thus on slower timescale than spiking discharges . We propose a mechanism around seizure onset: the frequency of inhibitory population spikes increases until the excitatory population starts to fire ( onset ) , and from there , inhibitory neurons reduce their activity to finally stop firing for the remainder of the event . Considering the temporal evolution of the slow variable in the deprived synaptic situation , we note that the inhibitory population’s activity is present when the variable is low . As soon as the excitatory lead takes place , the slow variable rises and quickly the inhibitory population becomes quiescent . This prediction would imply that the action of the excitatory activity on the extracellular medium acts as brake on synchronized inhibitory population’s spikes . Running systematic simulations over coupling strength parameters , we also observe that seizure duration decreases as we increase inter-population coupling ( Fig 8 , left ) . This can be understood as follows: as inter-population coupling increases , the synaptic gain of GABA synapses coming from synchronized spiking of inhibitory cells increases , which reduces the activity in the excitatory cells , eventually leading to the destruction of the self-sustained epileptiform discharges . It is notable that the modification of inter-population coupling results in multi-scale effects on fast and slow time scales and , in particular , significantly affects slower time-scales such as seizure duration . We finally performed simulations with different noise levels , as noise has a strong effect on neuronal network dynamics [41 , 42] . Our results support the idea that a noisy environment impacts epileptiform activity . In the context of epilepsy , the degree of stochasticity versus determinism in neural systems has been analyzed , but evidence for one or the other mechanism is still lacking [43] . With our model , low noise intensity in the spontaneous seizure configuration leads to long lasting , sharp onset and offset seizures with long periods between seizure-like activities . Increasing noise intensity results in more frequent but shorter seizure-like discharges ( Fig 8 , right ) , with less synchronized ensembles of neurons .
Approaches using dynamical system theory may provide guidance when deriving biophysical models of brain function and dysfunction . Here we used the abstract Epileptor model to guide the design of population models of coupled excitatory and inhibitory spiking neurons , which now allows studying the organization of spiking patterns as a function of coupling and excitability . We demonstrated that gap junction coupling plays the dominant role in synchronizing both neuron types , whereas the slow permittivity changes act rather as a slowly changing control parameter aiding in organizing the seizure progression . Our simulations support that the large amplitude discharge of spike-wave components of interictal and ictal population spikes arises from synchronized discharges of inhibitory interneurons only , which is not in line with current thinking in the literature , though finds support by recent empirical studies . Our approach does not rule out other physiological organizations giving rise to the same dynamics as described in the Epileptor; in fact , there is a large range of candidates . Thus although our findings may have validity only within a small range of physiological realizations , they nevertheless can give insight about experimental paradigms via simulations and analyses bridging the gap between neuronal spiking , network and abstract seizure evolution across large temporal scales . | Neurons communicate via different types of synapses on very fast time scales . The combination of hundred thousand of such interconnected cells within a fluctuating extracellular environment forms a complex network that gives rise to function and behavior via the formation of dynamical patterns of activity . In the context of epilepsy , the functional properties of the network at the source of a seizure are disrupted by a possibly large set of factors at the cellular and molecular levels . It is therefore needed to sacrifice some biological accuracy to model seizure dynamics in favor of macroscopic realizations . Here , we present a neuronal network model that convenes both neuronal and network representations with the goal to describe brain dynamics involved in the development of epilepsy . We compare our modeling results with animal in vivo recordings to validate our approach in the context of seizures . Such system-level methodology has significant bearing in understanding neuronal network dynamics that entangle multiple synaptic and extracellular modalities . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Computational Modeling of Seizure Dynamics Using Coupled Neuronal Networks: Factors Shaping Epileptiform Activity |
Heterodimers of CLOCK and BMAL1 are the major transcriptional activators of the mammalian circadian clock . Because the paralog NPAS2 can substitute for CLOCK in the suprachiasmatic nucleus ( SCN ) , the master circadian pacemaker , CLOCK-deficient mice maintain circadian rhythms in behavior and in tissues in vivo . However , when isolated from the SCN , CLOCK-deficient peripheral tissues are reportedly arrhythmic , suggesting a fundamental difference in circadian clock function between SCN and peripheral tissues . Surprisingly , however , using luminometry and single-cell bioluminescence imaging of PER2 expression , we now find that CLOCK-deficient dispersed SCN neurons and peripheral cells exhibit similarly stable , autonomous circadian rhythms in vitro . In CLOCK-deficient fibroblasts , knockdown of Npas2 leads to arrhythmicity , suggesting that NPAS2 can compensate for loss of CLOCK in peripheral cells as well as in SCN . Our data overturn the notion of an SCN-specific role for NPAS2 in the molecular circadian clock , and instead indicate that , at the cellular level , the core loops of SCN neuron and peripheral cell circadian clocks are fundamentally similar .
Endogenous circadian ( ca . 24 h ) biological clocks have evolved in virtually all organisms in order to synchronize physiological processes and behavior with 24 h day/night cycles and to anticipate reliably recurring daily events . In mammals , the master circadian clock is located in the brain , in the hypothalamic suprachiasmatic nucleus ( SCN ) [1] . The SCN receives light information directly from the retina via the retino-hypothalamic tract , allowing it to synchronize to external light/dark cycles . Through direct and indirect signals , the SCN regulates subsidiary circadian clocks in peripheral tissues that coordinate rhythms of local physiological processes [2] . At the cellular level , the circadian clock is comprised of clock genes that interact in transcriptional-translational delayed negative feedback loops ( TTLs ) [3 , 4] . In the core TTL that is essential for rhythmicity , BMAL1 and CLOCK form heterodimers that activate the transcription of Period ( Per1 , 2 , 3 ) and Cryptochrome ( Cry1 , 2 ) genes . PER and CRY then form complexes that translocate into the nucleus and suppress the activity of BMAL1:CLOCK , thereby inhibiting their own transcription . When PER/CRY complexes are eventually degraded in a controlled fashion , this inhibition is relieved , BMAL1:CLOCK heterodimers become active again , and the ca . 24 h cycle begins anew . The first known mammalian core clock gene was identified in a behavioral screen of chemically mutagenized mice , which yielded the Clock-Δ19 mutant [5] . This mouse carries a mutation in intron 19 of the gene Clock , leading to expression of a non-functional CLOCK protein that competes with wild-type CLOCK for binding in molecular complexes . In constant darkness , heterozygous and homozygous Clock Δ19 mice show long free-running periods of circadian locomotor activity rhythms , and homozygotes generally become arrhythmic , although this varies with genetic background . Consistent with the behavioral phenotype , molecular rhythm amplitudes in the SCN are severely reduced in homozygous Clock-Δ19 mice , and peripheral tissues do not exhibit autonomous circadian oscillations [6–8] . These findings led to the initial assumption that CLOCK , like BMAL1 , is an essential component of the TTL , such that circadian clocks are non-functional in the absence of CLOCK [9] . However , development of a mouse with a complete knockout of the Clock gene changed this concept . In contrast to Clock-Δ19 mice , in constant darkness Clock-/- mice show strong free-running rhythms that have a period ~20 min shorter than wild-type and do not become arrhythmic [10–12] . Although molecular rhythms in the SCN and liver of Clock-/- mice are reduced in amplitude , they are stably expressed in animals kept in constant darkness . Conceivably , an unknown dimerization partner for BMAL1 , which can substitute for CLOCK in Clock-/- mice but unsuccessfully competes with the mutated CLOCK protein in Clock-Δ19 mice , is capable of maintaining circadian rhythms in the total absence of CLOCK protein . This unknown partner was identified as NPAS2 , a paralog of CLOCK that dimerizes with BMAL1 to form transcriptionally active complexes [13] . Whereas knockout of the Npas2 gene has no severe effects on behavioral rhythms in mice , double-knockout of Npas2 and Clock leads to total arrhythmicity of behavior and of SCN clock gene expression . Interestingly , however , NPAS2 has been shown to have a role in controlling circadian rhythms only in the SCN , where NPAS2 can substitute for CLOCK , and in the forebrain , where NPAS2 is required for rhythmicity [13 , 14] . Even though NPAS2 is expressed in peripheral tissues and is in fact upregulated in livers of Clock-/- mice [10] , isolated peripheral tissues from Clock-/- mice , no longer under the influence of the SCN , were reported to lack circadian rhythms [15] . Together , these results led to the current concept that NPAS2 can compensate for the loss of CLOCK in the SCN but not in peripheral tissues [9] . Here , we show that peripheral fibroblasts from Clock-/-; mPer2Luc mice exhibit sustained , cell-autonomous circadian PER2 oscillations of similar quality to those of dispersed SCN neurons , which however were much less robust than those of SCN neurons in slice cultures , in which the reinforcing effects of SCN neuronal network interactions are largely preserved . Various peripheral tissues of Clock-/- mice exhibited sustained , autonomous circadian rhythms , but with lower amplitude . Importantly , Npas2 knockdown rendered Clock-/- fibroblasts arrhythmic , demonstrating that NPAS2 can substitute for CLOCK to maintain rhythmicity in peripheral cells . Our data indicate that the role of NPAS2 in circadian clock function is similar in SCN and peripheral tissues .
Clock-/- mice show relatively stable circadian rhythms in locomotor activity [10] , an output of the circadian system that is directly controlled by the SCN . One possible explanation for this is that in SCN neurons NPAS2 can substitute for CLOCK in dimers with BMAL1 , such that transcriptional activation at E-boxes and circadian oscillations are maintained [13] . Alternatively , specialized coupling among component cellular oscillators within the SCN neuronal network increases robustness against mutations of several clock genes [16 , 17] , and this might also explain the preservation of locomotor activity rhythms in Clock-/- mice . To investigate autonomous circadian clock function in SCN and peripheral cells and tissues of Clock-/- mice , we crossed them with the mPer2Luc reporter line and obtained Clock-/- mice bearing the bioluminescent PER2 reporter . Similar to results previously reported for whole Clock-/- SCN explants [13] , single cells of cultured organotypic SCN slices of Clock-/- mice displayed very stable circadian rhythms that were almost indistinguishable from those of wild-type SCN cultures ( Figs 1A and 1B and S1A and S1 Dataset ) . Consistent with the behavioral phenotype , circadian periods of SCN cells of organotypic cultures from Clock-/- mice were shorter than for wild-type mice . However , rhythm amplitude , goodness of fit , and number of rhythmic cells were unaffected by the mutation ( Fig 1C ) . In contrast , when SCN cells were dissociated and studied in dispersed cultures lacking the specialized neuronal network that maintains synchrony of cells in SCN slices , the quality of their rhythms was reduced drastically ( Figs 2A and 2B and S1B and S2 Dataset ) . Dispersed SCN cells from Clock-/- mice exhibited not only shorter and less consistent circadian periods compared to wild-type cells , but also significantly reduced amplitudes , reduced goodness of fit , and a tendency toward smaller number of rhythmic cells ( Fig 2C ) . For comparison , we re-analyzed data obtained previously from Bmal1-/- SCN explants and dispersed SCN cells [18] using the same criteria used here for wild-type and Clock-/- cultures . As previously reported [18] , the SCN neuronal network partly compensated for the loss of BMAL1 ( Fig 1C ) , whereas dispersed Bmal1-/- SCN cells were mainly arrhythmic , and the few rhythmic dispersed Bmal1-/- SCN cells had extremely low amplitudes ( Fig 2C ) . Compared to Bmal1-/- SCN cells , Clock-/- SCN cells showed much stronger rhythms ( Figs 1C and 2C ) . Liver and lung tissues of Clock-/- mice were previously shown to fail to exhibit circadian rhythms when isolated from the SCN [15] . To examine circadian clock function of individual peripheral cells from Clock-/- mice , we assessed mPer2Luc expression of primary fibroblasts in dissociated culture . Unexpectedly , Clock-/- fibroblasts displayed relatively stable circadian rhythms ( Figs 3A and 3B and S1C and S3 Dataset ) . As for dispersed SCN neurons , Clock-/- fibroblasts exhibited shorter and less consistent circadian periods ( Fig 3C ) . Furthermore , rhythm amplitudes and goodness of fit were significantly reduced compared to wild-type fibroblasts , and fewer cells were rhythmic . We cultured organotypic tissue slices of liver , lung , kidney , and adrenal from Clock-/- mice and monitored tissue-autonomous mPer2Luc rhythms at baseline and after administration of 10 μM forskolin to the culture medium . Surprisingly , all four tissues exhibited distinct circadian rhythms at baseline , and oscillations became more pronounced after treatment with forskolin Figs ( 4A and S2 and S4 Dataset ) . However , there were differences in quality of circadian rhythms between wild-type and Clock-/- tissues and also among different tissues of the same genotype . At baseline , only Clock-/- lung and kidney showed the short circadian period characteristic of Clock-/- mice , whereas Clock-/- liver and adrenal showed periods comparable to those of wild-type tissues ( S3 Fig ) . However , after the forskolin treatment , only Clock-/- liver exhibited shorter periods ( Fig 4B ) . In all Clock-/- tissues , rhythm amplitude was significantly lower than in wild-type tissues ( Figs 4B and S3 ) . In addition , damping of rhythms of Clock-/- tissues , likely reflecting progressive desynchrony among cells , was faster than in wild-type tissues ( Figs 4B and S3 ) . Clock-/- and wild type tissues also differed in phase of circadian rhythms before and after resetting with forskolin ( Figs 4 and S3 ) . Although we found rhythms in all four peripheral Clock-/- tissues , not all individual explants showed significant rhythms . Without further synchronization , only 50% of Clock-/- liver explants and 25% of Clock-/- kidney explants were rhythmic , whereas all wild-type tissue explants and almost all Clock-/- lung and adrenal explants showed significant rhythms ( S3 Fig ) . Forskolin was able to induce rhythms in slices that had been arrhythmic before the treatment . After administration of forskolin , all Clock-/- liver and adrenal explants , 75% of Clock-/- lung explants , and 50% of Clock-/- kidney explants were rhythmic ( Fig 4B ) . Although rhythm quality and proportion of rhythmic explants was lower in Clock-/- tissues than in wild-type tissues , our findings reveal that peripheral tissues of Clock-/- mice are clearly capable of generating self-sustained circadian rhythms in the absence of any SCN input signal . To test potential changes of circadian clock gene expression patterns in vivo , we collected liver , lung , kidney , and adrenal tissues from wild-type and Clock-/- mice at 6 different time points throughout the day and examined mRNA levels of Npas2 , Bmal1 , Per2 and various clock controlled genes with qPCR . In Clock-/- livers and lungs , all genes showed significant circadian oscillations ( S4 Fig and S1 Table ) . However , in Clock-/- kidneys only Bmal1 expression was significantly rhythmic , and in Clock-/- adrenals all core clock genes , but not the clock controlled gene Star showed significant rhythms . Mean Npas2 expression was drastically increased in Clock-/- livers ( S4 Fig and S1 Table ) . Similar but less pronounced effects on Npas2 expression were seen in Clock-/- lungs and adrenals . Interestingly , although all mice were entrained to the same LD cycle , almost all investigated genes of Clock-/- tissues showed phase differences relative to wild-type tissues ( S4 Fig and S1 Table ) . In the SCN and for rhythms of locomotor activity , the CLOCK paralog NPAS2 can compensate for the absence of CLOCK in Clock-/- mice [13] . Since peripheral Clock-/- cells and tissues also exhibit autonomous circadian rhythms , we asked whether NPAS2 plays a similar role in peripheral cells . We cultured primary wild-type and Clock-/- fibroblasts and used lentiviral vectors expressing shRNAs to stably suppress Npas2 expression . The average efficiency of the Npas2-knockdown was 60% ( S5 Fig ) . Importantly , not all cells within a single culture dish were transduced by the viruses , allowing comparison of rhythms from affected and unaffected cells in the same dish , from the same mouse ( Fig 5A ) . Most Clock-/- fibroblasts with suppressed Npas2 expression were arrhythmic , whereas most fibroblasts from the same Clock-/- animals were rhythmic when Npas2 expression was not affected ( Fig 5B and 5C and S5 Dataset ) . In wild-type ( Clock+/+ ) fibroblasts , knockdown of Npas2 had no effect on the proportion of rhythmic cells , and infection with control viruses carrying a scrambled shRNA sequence did not change the proportion of rhythmic cells in either wild-type or Clock-/- fibroblasts ( Fig 5B ) . These findings suggest that NPAS2 is able to compensate for the loss of CLOCK by rescuing circadian rhythmicity in peripheral cells as well as the SCN .
Previous work developing and characterizing Clock-/-mice demonstrated that CLOCK is not essential for circadian oscillations of SCN clock gene expression or behavioral rhythms [10] . NPAS2 was identified as a substitute for CLOCK [13] , rescuing circadian rhythmicity in its absence , but only in the SCN central clock and not in the periphery , where tissues failed to produce autonomous rhythms in the absence of CLOCK ( and absence of SCN signals ) in vitro [15] . In contrast , our data reveal that peripheral tissues and single fibroblasts of Clock-/- mice are , in fact , capable of generating stable , autonomous circadian rhythms in vitro , albeit with diminished amplitude , reliability , and resetting capacity . Our experiments indicate that when cultured in the absence of SCN input signals , liver , lung , kidney , and adrenal explants from Clock-/- mice can exhibit stable circadian mPer2Luc oscillations . However , the period and phase of these rhythms are different from wild-type controls , and these genotypic differences vary across tissues . Clock-/- mice display a shorter free-running period of locomotor activity rhythms in constant darkness [10]; in our experiments , a corresponding shortening of circadian period is observed in SCN and in some ( but not all ) peripheral tissues cultured from Clock-/- mice . Also , Clock-/- mice show advanced phases of activity rhythms and enhanced phase shifting responses to light [11] . In cultured tissues and in vivo , the absence of CLOCK also affects phases of rhythmic expression of other core clock genes in a tissue-specific manner , suggesting a tissue-specific role for CLOCK in phase-setting and/or tissue-specific compensation mechanisms for the loss of CLOCK . Compared to wild-type tissues , rhythms of Clock-/- tissues generally exhibit lower amplitude , damp faster , and are less reliable . For example , when tested without prior synchronization by forskolin , only about 50% of Clock-/- liver explants are rhythmic , and even when rhythms are present , they are of lower amplitude and damp faster than rhythms in wild-type liver explants . This may explain why rhythms in Clock-/- peripheral tissues were not detected in a previous study by DeBruyne et al . , although they observed low-amplitude rhythmicity in raw data of lung mPer2Luc expression [15] . Thus , it is possible that different analysis methods or criteria for determining arrhythmicity explain the discrepancy between the studies . Technical differences may also contribute to the different results . For example , DeBruyne et al . used mice that were backcrossed to the C57BL/6J strain for 8–10 generations and were mPer2Luc heterozygous , whereas our mice were on a mixed background of C57BL/6J and Sv/129 , and mPer2Luc homozygous [15] . Finally , slight differences in culture procedures or media composition might differentially synchronize or amplify rhythms in vitro . Even simple changes of culture medium , for example , can temporarily synchronize cellular circadian oscillators , thereby revealing rhythmicity in an otherwise asynchronous population of peripheral cells [19] . In any case , whereas technical considerations could obscure rhythmicity , they cannot create rhythmicity de novo , so it seems safe to conclude that Clock-/- peripheral cells and tissues are , in fact , capable of autonomous circadian oscillation . Single-cell fibroblast experiments suggest that the weaker rhythms in Clock-/- tissues are due to both lower single-cell amplitude and less consistent circadian period . However , some tissues compensate better for the lack of CLOCK than others . In the SCN , the neuronal network preserved in SCN slices compensates very effectively for the loss of CLOCK ( Fig 1 ) , as it does for other clock gene defects [16 , 17] . In the periphery , Npas2 mRNA expression is more strongly upregulated in liver and adrenal than in lung or kidney , and this may account for the better rhythm amplitude and reliability in those Clock-/- tissues relative to corresponding wild-type tissues ( S4 Fig ) . The Clock-/- kidney , which shows the least compensatory upregulation of Npas2 expression of any of the four peripheral tissues studied , also shows the least reliable mPer2Luc expression among explants in vitro ( Figs 4 and S3 ) , and the least consistent rhythms of gene expression in vivo ( S4 Fig ) . However , even strong upregulation of Npas2 in the liver does not fully substitute for CLOCK . The loss of CLOCK also leads to a decrease of BMAL1 [10] , which is the binding partner of NPAS2 . Thus , without sufficient levels of the binding partner BMAL1 , even high amounts of NPAS2 may not be able to compensate completely for the loss of CLOCK . Although intrinsic differences of molecular clock mechanism may exist across cell and tissue types , e . g . level of compensatory upregulation of Npas2 expression , and these differences may contribute to the differing effects of Clock knockout that we observed across peripheral tissues , we found no evidence for such differences between SCN neurons and fibroblasts . In fact , effects of Clock knockout on cellular rhythmicity are remarkably similar in dispersed SCN neurons and fibroblasts . In both Clock-/- and wild-type cells , rhythm amplitude is greater in SCN neurons than in fibroblasts . However , in both types of cells , amplitude , goodness of fit , and the proportion of rhythmic cells are all similarly reduced in the absence of CLOCK ( Figs 2C and 3C ) . Consistent with this , Clock-/- mice appear to accumulate BMAL1 in only ~10% of SCN neurons , suggesting a lower proportion of competent circadian oscillators in the SCN of these mice [10] . Together , this demonstrates that the previously observed robustness of SCN slice rhythms ( and behavioral rhythms ) to Clock knockout [10 , 13] can be explained by the specialized coupling within the SCN neuronal network that is lacking in peripheral tissues [16 , 17] . Earlier studies by DeBruyne et al . of tissues from double knockout mice deficient in both CLOCK and NPAS2 indicated that NPAS2 can substitute for CLOCK to maintain circadian rhythmicity in SCN but not in peripheral tissues [15] . However , PER2 rhythms in forebrain are abolished in NPAS2-deficient mice [14] , implying an important role for NPAS2 in non-SCN clocks . Other studies have also suggested a role for NPAS2 in peripheral circadian clocks , although rhythms were not investigated in those studies [20–22] . NPAS2 binds to BMAL1 in whole brain and in peripheral tissues [10 , 21] and to E-boxes in the mouse liver [3] , suggesting that it could be important for rhythmic transcription in peripheral cells . Importantly , we show here that knockdown of NPAS2 in Clock-/- fibroblasts leads to complete arrhythmicity in most cells , thus establishing that NPAS2 can compensate for loss of CLOCK in peripheral cells as well as in SCN . In summary , our results demonstrate that CLOCK is not essential for circadian clock function in peripheral tissues , because NPAS2 can substitute for CLOCK in these tissues , as it can in the SCN . However , this genetic compensation is incomplete , resulting in weaker rhythms in peripheral tissues of Clock-/- mice . Neuronal network interactions specific to SCN further compensate for absence of CLOCK ( as previously observed for other clock gene defects [16 , 17] ) , resulting in nearly normal circadian rhythms in SCN slices in vitro or locomotor activity in vivo [10 , 11 , 13] . Rhythmicity of gene and protein expression in peripheral tissues has previously been reported in Clock-/- mice [10] , but this might reflect SCN-driven oscillations , whereas our in vitro results establish that Clock-/- peripheral tissues are capable of truly autonomous circadian oscillation . A further specific implication of our results , consistent with previous work but at times underappreciated , is that Clock-/- mice are not appropriate for experimental studies aiming to eliminate peripheral circadian clock function , because peripheral cells from these mice are in fact rhythmic; instead , such studies should use Bmal1-/- or Cry1-/-;Cry2-/- mice , as lack of rhythmicity has been demonstrated in peripheral cells from these mice [16 , 18] . Finally , our in vivo results demonstrate tissue-specific differences in the role of NPAS2 in the circadian clock , and further studies are needed to elucidate these tissue-specific mechanisms .
Mouse studies were approved by the Institutional Animal Care and Use Committee at University of California , San Diego ( Protocol number: S07365 ) . Every effort was made to minimize the number of animals used , and their suffering . Clock-/- mice [10] were crossed to mPer2Luc -SV40 reporter mice [19 , 23] . From the heterozygous offspring , we created a double homozygous Clock-/-; mPer2Luc and a Clock+/+; mPer2Luc line , both with the same mixed genetic background . Henceforth , for convenience , these mice will be referred to as Clock-/- and WT . All experiments for this study were carried out in 2–4 month old male mice , except SCN cultures which were from neonatal ( 3–6 day old ) male and female mice . Mice were maintained in LD 12:12 cycles ( 12 h light , 12 h dark , lights on at 06:00 hr ) with ad libitum access to food and water . Primary fibroblasts were obtained from tail or ears . Tissues from each mouse were chopped into small pieces with a scalpel and incubated twice in a 0 . 25% trypsin solution for 30 min at 37°C . The tryspin digestion was stopped by adding culture medium ( high glucose , pyruvate DMEM [Gibco , #11995] with 5% FBS and 50 U/ml penicillin , 50 μg/ml streptomycin ) . Cells were centrifuged ( 0 . 5 x g , 5 min , room temperature ) and resuspended in fresh culture medium . About 2x105 cells/dish were seeded onto 35 mm dishes and incubated at 37°C with 5% CO2 . When cells reached ~70% confluence , culture medium was replaced with explant medium formulated for equilibration with 5% CO2 ( high glucose DMEM [Mediatech , Manassas , VA , USA] , 14 mM sodium carbonate , 10 mM HEPES , 52 U/ml penicillin , 52 μg/ml streptomycin , 4 mM L-glutamine , 2% B-27 [GIBCO , Grand Island , NY , USA] , 0 . 1 mM luciferin [BioSynth , Itasca , IL , USA] ) . Tissues were isolated and kept in half-frozen Hank’s Balanced Salt Solution ( HBSS ) . 300 μm brain slices were prepared with a vibratome ( Leica VT1200S , Buffalo Grove , IL , USA ) . Tissues were immediately transferred to tissue culture inserts ( EMD Millipore , Billerica , MA , USA ) and cultured in 35 mm dishes containing 1 ml of explant medium formulated for equilibration with air ( high glucose DMEM [Mediatech , Manassas , VA , USA] , 4 mM sodium carbonate , 10 mM HEPES , 52 U/ml penicillin , 52 μg/ml streptomycin , 4 mM L-glutamine , 2% B-27 [GIBCO , Grand Island , NY , USA] , 0 . 1 mM luciferin [BioSynth , Itasca , IL , USA] ) . When indicated , tissues were treated with 10 μM forskolin for 2 hours in order to enhance rhythmicity . Luminescence measurements were taken at 10 min intervals using a LumiCycle luminometer ( Actimetrics ) that was placed inside a 37°C incubator without CO2 . Period , peak phases , goodness of fit , and amplitude were determined over 7 days by fitting a sine wave [Sin fit ( Damped ) for period , phase , and damping constant ( days to reach 1/e of initial amplitude ) , or LM fit ( Sin ) for amplitude] to 24 h running average baseline-subtracted data using LumiCycle Analysis software ( Actimetrics , Wilmette , IL , USA ) . The first day of measurement was excluded from analyses . Amplitude was normalized to total brightness in order to account for different sizes of brain tissue and technical differences between slices . Explants failing to show significant χ2 periodogram values near 24 h [24] , a goodness of fit >0 , or a minimum of two mPer2Luc peaks were determined to be arrhythmic and were excluded from further quantification . Single-cell mPer2Luc measurements were carried out as described elsewhere [25 , 26] . Briefly , a sealed culture in air-equilibrated explant medium was placed on the stage of an inverted microscope ( Olympus IX-71 , Tokyo , Japan ) in a dark , windowless room . A heated lucite chamber , custom-engineered to fit around the microscope stage ( Solent Scientific , Segensworth , UK ) , kept the sample at a constant 36°C . Light from the sample was collected by an Olympus 4x XLFLUOR objective ( NA 0 . 28 ) and transmitted directly to a cooled charge-coupled-device ( CCD ) camera ( Spectral Instruments , Tucson , AZ , USA ) mounted on the bottom port of the microscope . The camera contained a back-thinned CCD thermoelectrically cooled to -90°C with a rated quantum efficiency of 92% at 560 nm . The signal-to-noise ratio was increased by 4 × 4 binning of the 1056 × 1032 pixel array . Images were collected at intervals of 30 min , with 29 . 5 min exposure duration , for 4–7 days . Images were acquired and saved to a computer with SI Image SGL D software ( Spectral Instruments ) , and analyzed with MetaMorph ( Molecular Devices , Sunnyvale , CA , USA ) . Period , goodness of fit , and amplitude were determined over 7 days by fitting a sine wave [Sin fit ( Damped ) for period , phase , and damping constant ( time to reach 1/e of initial amplitude ) , or LM fit ( Sin ) for amplitude] to 24 h running average baseline-subtracted data using LumiCycle Analysis software ( Actimetrics , Wilmette , IL , USA ) . Cells failing to show significant χ2 periodogram values near 24 h [24] , a goodness of fit >0 , or a minimum of two mPer2Luc peaks were determined to be arrhythmic and were excluded from further quantification . Quantitative real-time PCR ( qPCR ) was performed with a CFX384 thermocycler system ( Bio-Rad , Hercules , CA ) with GoTaq SYBR Master Mix ( Promega , Madison , WI ) . Relative quantification of expression levels by a modified ΔΔCT calculation was performed as described [27] . ß-Actin was used as a reference gene . PCR primer sequences are listed in S2 Table . Primary fibroblasts were transfected with either Npas2 shRNA ( GCTCCGAGAATCGAATGTGAT and GCAAGAACATTCCGAAGTTTA ) or scrambled control shRNA ( TCGTTTACCACCTCCTGCA ) lentiviral particles that also contained a GFP marker sequence . Cells were split and grown until ~70% confluency was reached ( max . 48 hours ) . Medium was reduced and polybrene ( 4 μg/ml ) was added . 10 μl of each Npas2 and 20 μl of the control virus stock solution were added and cells were incubated for 3 hours at 37°C . Afterwards medium was changed . Seven days after transfection , cells were used for single cell mPer2Luc measurements . The efficiency of the Npas2 knockdown was tested with qPCR after cells were sorted by flow cytometry according to their expression of GFP ( S5 Fig ) . Statistical analyses were conducted using GraphPad Prism . Rhythmicity and phase of qPCR clock gene expression profiles were determined using CircWave v1 . 4 ( developed by Roelof Hut , University of Groningen , Netherlands , http://www . euclock . org/results/item/circ-wave . html ) . CircWave v1 . 4 fits data to a sine wave with added harmonics . Significance of the curve fit is tested against a fitted horizontal line through the overall average . Phase is calculated as the Center of Gravity of the fitted wave form . For raster plots , bioluminescence intensity values were normalized to mean intensity for each cell . Using Gene Cluster 3 . 0 and Treeview ( developed by Dr . Michael Eisen while at Stanford University , USA ) , data were color coded , with green for positive and red for negative values . Details about statistical tests used for individual experiments are indicated in the figure legends . | In mammals , circadian clocks are based on a core transcriptional–translational feedback loop . BMAL1 and CLOCK activate the transcription of Per1-3 and Cry1/2 . PER and CRY proteins inhibit BMAL1/CLOCK , and thus their own transcription . In Clock-/- mice , NPAS2 can substitute for CLOCK in the suprachiasmatic nucleus ( SCN ) , the major circadian pacemaker . However , peripheral tissues of Clock-/- mice were reported to lack circadian rhythms . Since then , the protein CLOCK has been deemed essential for circadian rhythms in peripheral tissues . However , here we show that Clock-/- peripheral cells and tissues exhibit stable , autonomous circadian rhythms . Furthermore , in Clock-/- fibroblasts , knockdown of Npas2 leads to arrhythmicity , suggesting that NPAS2 can compensate for the loss of CLOCK in peripheral cells as well as in SCN . Our data overturn the notion of an SCN-specific role for NPAS2 , and instead indicate that the core loops of SCN neuron and peripheral cell circadian clocks are fundamentally similar . This finding redefines a basic principle of molecular circadian clock regulation in peripheral organs which are essential for many metabolic processes . Disturbances of these rhythms lead to disorders like diabetes , obesity , and cancer . Thus , understanding the molecular basis of peripheral circadian oscillators is essential to develop treatments against clock-related disorders . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"vertebrates",
"mice",
"animals",
"electromagnetic",
"radiation",
"fibroblasts",
"circadian",
"oscillators",
"mammals",
"connective",
"tissue",
"cells",
"chronobiology",
"kidneys",
"animal",
"cells",
"connective",
"tissue",
"biolog... | 2016 | NPAS2 Compensates for Loss of CLOCK in Peripheral Circadian Oscillators |
New efforts are being made to improve understanding of the epidemiology of the helminths and intensifying the control efforts against these parasites . In contrast , relatively few studies are being carried out in this direction for the intestinal protozoa . To contribute to a better comprehension of the epidemiology of the intestinal protozoa , prevalence , and spatial distribution of Entamoeba histolytica/dispar and Giardia lamblia , and their association with drinking water supplies , were determined in the Agboville department in southeast Côte d'Ivoire . Stool samples were taken from more than 1 , 300 schoolchildren in the third year of primary education ( CE1 ) from 30 primary schools and preserved in SAF ( sodium acetate-acetic acid-formalin ) . The samples were analyzed by formalin-ether concentration . Then , a survey questionnaire addressed to schoolchildren and school directors was used to collect data on water supplies . Prevalence of E . histolytica/dispar and G . lamblia were , respectively , 18 . 8% and 13 . 9% . No particular focus zone was observed in the spatial distribution of the two species . Significant negative association was observed between use of tap water and high prevalence of E . histolytica/dispar infection ( OR = 0 . 83 , p = 0 . 01 ) . High prevalence of G . lamblia infection was positively associated with use of ponds as the source of drinking water ( OR = 1 . 28 , p = 0 . 009 ) . These two species of pathogenic protozoa are present with substantial prevalence in this area of Côte d'Ivoire . Although their spatial distribution is not focused in any one place , determination of the population segments with the highest levels of infection will help to target the chemotherapeutic fight . To reinforce treatment with chemotherapeutic agents , tap water should be made available in all the localities of this area .
Although intestinal parasites seem to raise much less interest than do AIDS and tuberculosis , they are a major public health problem in tropical regions [1] . In 2002 , WHO estimated the number of people infected by digestive tract parasites at 3 . 5 billion and the number of people made ill by them at 450 million [2] . Whereas much effort is being made toward a better comprehension of helminth epidemiology [3] , [4] , relatively few equivalent studies are done on intestinal protozoa . This is surprising , because intestinal amebiasis caused by the protozoan Entamoeba histolytica is the third-greatest parasitic disease responsible for death in the world after malaria and schistosomiasis [5] , [6] . It affects approximately 180 million people , of whom 40 , 000 to 110 , 000 die each year [7] . Giardiasis , caused by Giardia lamblia , is a frequent cause of diarrhea [8] , [9] that can have a negative impact on growth and development of children [10] and affects approximately 200 million people worldwide [11] . These parasitic diseases are found in all the major regions of Africa [12]–[14] and were reported in Côte d'Ivoire by surveys carried out in the west of the country [15]–[18] . Giardia cysts were reported in an investigation on an epidemic of diarrhea that occurred in the village of Offoumpo in Agboville area [19] . Other studies in this area also reported a high prevalence of certain protozoal species such as E . histolytica [20] . In the same area , N'Guessan et al . found that the very high rate of blood in feces is associated with intestinal schistosomiasis [21]; these authors thought that blood in feces could also be due to other diseases such as amoebiasis . A parasitological survey should help to establish the existence of amoebiasis and assess the probable contribution of E . histolytica to the occurrence of fecal blood in the Agboville area . Treatment of giardiasis and intestinal amoebiasis relies on derivatives 5-nitro-imidazoles such as the metronidazole , marketed since 1959 [22] . To date , some resistant cases of G . lamblia to these products have been reported . Unfortunately , no new drug is under development for specific treatment of intestinal protozoa [22] . In order to reduce or delay development of resistance , certain authors recommend avoidance of mass treatments in favor of targeted treatments and greater effort put into prevention [22] , [23] . Collection of epidemiological data is necessary to develop fight effective strategies against these parasites . The main objective of this study was to estimate the prevalence of intestinal protozoa in the feces of schoolchildren in the Agboville area . The secondary objectives were to establish spatial distribution of E . histolytica and G . lamblia in this area and to determine the relationship between these parasites and household water sources . The results should facilitate evaluation of the endemic level of these parasites and to know if infection risk is focused in an area or is widely spread , and consequently whether massive or focal measures of parasite control are required .
The study was carried out in the Agboville area , southeast Côte d'Ivoire ( 3°55′ and 4°40′ West and 5°35′ and 6°15′ North ) . The area is rugged and consists of numerous valleys with swamps . It is a forested region and the climate is of equatorial type with two rainy seasons and two dry seasons . Its average annual rainfall is between 1 , 298 and 1 , 739 mm with temperature ranging between 25 and 26 . 6°C [24] . This zone covered by a dense hydrographic network made up of two rivers ( Agnéby and Mé ) . The tributaries and streams are numerous and conduct water to some villages; there are also many isolated rivers . The Agboville area has 103 villages and the population is estimated at 244 , 865 people , most of whom are farmers . The main crops are cocoa , coffee , and food products as in west of the county . Populations are supplied with water by traditional dug wells , boreholes , taps ( which are supplied by wells or public water delivery systems ) , rivers , and ponds . The study population consisted of schoolchildren . The study presented here used two surveys: first , a comprehensive parasitological survey in all the primary schools of one education inspection in the Agboville area that fulfilled our inclusion criteria ( i . e . , that they were registered in one of the schools of the Agboville inspection ) ; second , a questionnaire survey to collect data on water sources . Institutional approval of the study protocol was granted by Abidjan-Cocody University ( IRB 09-2003 ) . The study received ethical clearance by the Ministry of Public Health in Côte d'Ivoire . Then we obtained the oral consent of teachers and parents of pupils according to the principles of the Declaration of Helsinki , before beginning the data collection . The consent was oral because the majority of the parents cannot read nor write . Documentation of this oral consent was initialed and dated by the examiner according to data collection forms approved by the IRB . It was also approved by the organization of parents of pupils . Participation of pupils was voluntary . Those who refused to give fecal samples or to answer the questionnaire were simply excluded from the study . At the end of the parasitological survey all schoolchildren were treated without cost with albendazole for soil-transmitted helminth infections and intestinal protozoa . The list of all 30 primary schools in the area was provided by the inspector . Then , in the last week of October 2004 , he explained the aim and the procedures of the study to the school directors and requested class lists with name , age , and sex of each pupil . Sampling was done from all voluntary schoolchildren of the third grade class of the inspection during the last two weeks of November 2004 . After the children were given an explanation of the stool sample process , they received plastic , covered 125 ml transparent tubes into which they placed their samples . Tubes were given labels to identify the sample , then placed in racks and transported to the laboratory of the major diseases of Agboville ( the state-run laboratoire des grandes endémies d'Agboville ) . To preserve the samples , 1 to 1 . 5 g of stool was placed ( by wooden stick ) in another tube containing 10 ml of sodium acetate–acetic acid–formalin ( SAF ) solution carrying the same label as the corresponding tube . Thereafter , tubes were shaken vigorously to mix feces and SAF [25] , [26] . SAF tubes were transported to the laboratory in Abidjan , where stool samples were analyzed by formol-ether concentration [27] and examined by microscopy . All the species of intestinal protozoa and helminths observed were recorded . Slides were read semiquantitatively for intestinal protozoa ( 1+ , 2+ , 3+ according to parasitic load of microscopic fields ) and quantitatively for helminths ( eggs were counted systematically ) . Two weeks before the parasitological survey , the questionnaire was distributed in all 30 schools . This questionnaire was used in previous studies in Côte d'Ivoire [28] , [29] . It takes into account other aspects but only data on water supply sources were considered for this study . It consists of two parts , one sent to school directors and the other to teachers . The part sent to school directors was filled in by them . The part sent to teachers was used to collect data from the students . The teachers followed the instructions that accompanied the questionnaire and interviewed pupils separately , one after another , in an empty classroom to avoid the influence of others on their responses . Answers to the questions were “O” for “yes” , “N” for “no” and “—“ for “I do not know . ” Completed questionnaires were collected during the parasitological investigations . During the parasitological investigation , geographical coordinates and altitude of each school were recorded with GPS ( global positioning system; Magellan 315 , Thales Navigation , Santa Clara , California , United States ) ; then information on roads were identified on-site and those on rivers were observed on maps . Data were used to develop geo-referenced files of the Agboville area from existing maps , ArcView ( Redlands , California , United States ) , and MapInfo . The prevalence of parasites ( by species ) was then incorporated into the digital map . Data were double entered and validated with EpiInfo 2002 ( US Centers for Disease Control and Prevention , Atlanta , Georgia , United States ) . Two age groups , 6–10 and 11–12 years , were performed . Chi-square ( χ2 ) tests were conducted with STATISTICA 6 . 0 ( StatSoft , Data Analysis Software System , Tulsa , Oklahoma , United States ) , to determine the relationship between parasites and the children's age and sex with a confidence interval ( CI ) of 95% . The relationship between the prevalence of different species of parasites was evaluated by the Pearson correlation coefficient ( r ) and its significance ( p-value ) by linear regression carried out with STATISTICA 6 . 0 . Associations between parasite prevalence and water supply sources were examined by logistic regression conducted with STATA 9 . 0 ( Stata , College Station , Texas , United States ) .
All 30 schools of the Agboville inspection participated in the study . Out of 1 , 500 schoolchildren who were registered on the class lists , 89 did not provide stool samples ( 27 were absent during the study and 62 refused to participate ) and 37 did not complete the questionnaire ( Figure 1 ) . Consequently , 1 , 398 schoolchildren ( 93 . 2% ) provided stool samples and answered the questionnaire . Only these were included in the analysis . Eight species of intestinal protozoa , including two pathogenic species , were found in the stool samples . E . histolytica/dispar was found in 263 pupils ( 18 . 8% ) ( Table 1 ) and G . lamblia was found in 195 ( 13 . 9% ) . 2 . 9% of the pupils were infected by both species and 29 . 7% were infected by at least one of them . In addition to these pathogenic species , six nonpathogenic species were found among the samples: Entamoeba hartmanni , Entamoeba coli , Endolimax nana , Iodamoeba butschlii , Chilomastix mesnili , and Blastocystis hominis . The most common species were E . nana and E . coli , with respective prevalence of 65 . 5% and 62 . 3% . Concerning the prevalence of protozoal infection by age and sex , we found a significant association between the prevalence of G . lamblia and sex ( χ2 = 7 . 32 , df = 1 , p = 0 . 006 ) and between the prevalence of C . mesnili and age ( χ2 = 4 . 25 , df = 1 , p = 0 . 037 ) . No other significant association with age and sex were found . We found that only 118 ( 8 . 4% ) students carried no protozoal species . However , among the infected schoolchildren , 326 ( 23 . 3% ) were infected by one protozoal species and the remaining children ( 68 . 2% ) had multiple infections . In the multiple-infection group , 2 . 5% were infected by E . histolytica/dispar , G . lamblia and E . coli and 12 . 3% by E . histolytica/dispar , E . nana , and E . coli . The five sites studied in the town of Agboville were so close that they merged into a single point on the map . Students in all the study sites in this area were infected by these two parasites ( Figure 2 ) . E . histolytica/dispar prevalence varied from 4 . 2% in Agboville town to 40 . 8% in Oress-Krobou; prevalence exceeded 20% in almost half of the study locations ( 13 of 30 ) . The villages with the greatest prevalence were distributed throughout the Agboville area . The prevalence of G . lamblia ranged from 2 . 0% in the village of Loviguié to 26 . 8% at Kouadjakro . Eight localities had prevalence that exceeded 20%: Séguié , Boka Oho , Kouadjakro , Babiahan , Ery-Makouguié , Grand Moutcho , Gbéssé , and Anno . These villages were also distributed throughout the area without focus zone . In addition , Seguié , Boka Oho , Kouadjakro , Ery-Makouguié , Grand Moutcho , and Gbéssé were the most infected localities by both pathogenic species . The spatial distribution of E . histolytica/dispar and the rate of blood in stools cannot be superimposed , even if in certain localities prevalence coincides . Localities where blood in stools is found in more than 20% of the schoolchildren were identified during a study carried out by Guéssan et al . [21] . In our study , prevalence of the three intestinal parasites species suspected to cause blood in stools was evaluated in these localities ( Table 2 ) . In Gbéssé , Séguié , and Offompo , only E . histolytica/dispar had a prevalence higher than 20% . In these locations , this species is likely the most responsible for the observed blood in stools . In Ery-Makouguié and Oress-Krobou , E . histolytica/dispar together with Ancylostoma spp . has a prevalence higher than 20%; both species could account for the observed rate of blood in stools . In Odoguié and Yadio , E . histolytica/dispar could have slightly contributed to blood in stools , given the higher prevalence of Ancylostoma spp . and S . mansoni . In all other localities the prevalence of E . histolytica/dispar is lower than 20% so this species seems not have contributed to blood in stools there . Water sources in the Agboville area consist of taps , dug wells , boreholes , rivers , backwaters ( shallows area adjacent to or part of a river , where clothing is often laundered ) , ponds , and reservoirs . Most people supply themselves with water from several sources at the same time . Dug wells are the main source in both towns and villages . A significant negative association was observed between use of tap water and a high prevalence of E . histolytica/dispar ( odds ratio OR = 0 . 84 , 95% CI = 0 . 73–0 . 96 ) . High prevalence of G . lamblia infection had a significant positive association with use of pond water ( OR = 1 . 28 , 95% CI = 1 . 06–1 . 53 ) . The results show also a significant positive association between high prevalence of non-pathogenic E . coli and E . nana with , respectively , backwaters and rivers . These two nonpathogenic intestinal protozoa had a significant negative association with use of tap water ( Table 3 ) .
The study was carried out among schoolchildren because they were one of age groups the most exposed to intestinal parasites and were generally accessible . Those of third grade primary school ( CE1 ) were chosen because they were the youngest pupils able to answer to questions without difficulty and could be followed over several subsequent years . The method used for stool analysis , formol-ether concentration [25]–[27] , did not allow a distinction between E . histolytica and E . dispar , so these parasites were indicated by E . histolytica/dispar . More specialized methods now exist to distinguish them [30] , [31] but remain inaccessible in the majority of developing countries [32] . The prevalence of this parasite complex in our study ( 18 . 8% ) is identical to that obtained by Heckendorn et al . in 2002 in the town of Agboville [20] . In addition , our extended areas of sampling show that beyond Agboville town , the parasite complex infects people in the wider area ( including villages ) beyond the town Agboville , and maintains its level of infection in the population . In the Man area , in west Côte d'Ivoire , prevalence of the complex E . histolytica/dispar is even lower , with a rate of 11 . 3% [33] . Distinction between the two species E . histolytica and E . dispar could led to a weaker prevalence of the pathogenic species [30] , [31] . In Agboville town , in an analysis of only microscopically positive samples by PCR , the ratio E . histolytica to E . dispar was 1∶46 [20] . On the basis of this ratio , prevalence of the pathogenic species ( E . histolytica ) in our study could be about 0 . 4% . However , studies have shown a significant association between this complex and diarrhea in Nigeria [32] , so the high prevalence of the E . histolytica/dispar complex as a contributor to illness must nevertheless be considered , even if it is controversial . Prevalence of G . lamblia in Agboville area was 13 . 9% . This is above other estimates for Côte d'Ivoire: 10 . 8% in the Man area [18] and 1 . 4% in Toumodi in central Côte d'Ivoire [34] . The higher prevalence of this parasite in the Agboville area could be due to higher rainfall [24] . Protozoal infection was associated with age and sex for two species . The 6- to 10-year age group was the most infected by C . mesnili . This has been observed in the west [17] and in other African countries [1] , [35] and is due to the risky behavior and relatively poorer hygiene measures in this age group . G . lamblia infection was associated with sex , with girls more highly infected . Where surface water is used for household activities , girls are more vulnerable as indicated by Brelet [36] . Concerning polyparasitism , our results are comparable to those of Keiser et al . obtained in western Côte d'Ivoire [17] . The observed multiple infections could be explained by the facts that many species of protozoa have the same mode of transmission and that hygiene is poor in these areas . E . histolytica/dispar and G . lamblia were found in samples from all the localities studied . This cosmopolitan distribution of these parasites has been reported by some authors [7] , [31] . Localities of high prevalence are distributed throughout the Agboville area . The even spatial distribution of E . histolytica/dispar is identical to that observed in the Man area in western Côte d'Ivoire . In contrast , three focal zones were observed in the spatial distribution of G . lamblia in the Man area , contrary to the Agboville area [33] . The even distribution of these parasites in the Agboville area shows that transmission is not related to the physical environment of the area but to the fact of specific parameters of each locality . Eradication efforts should thus take into account the entire area without stratification and emphasize improvements in hygiene conditions . Among the 13 localities with E . histolytica/dispar prevalence over 20% , Offompo , Grand Moutcho , Oress Krobou , Yadio , Ery-Makouguié , Odoguié , and Gbéssé also have a fecal blood rate of over 20% [21] . This blood may be due to S . mansoni or Ancylostoma spp . , but E . histolytica/dispar is likely to be a cause only in Séguié , Offompo , and Gbéssé . In Azaguié and Agboville , the prevalence of pathogenic protozoan infection is low , as is the blood rate in stool , certainly because these localities benefit from a distribution network of safe drinking water and hygiene conditions are better . The socioeconomic status of the populations has not been taken into account in this study . However , recent work has shown that income levels of people influence the distribution of intestinal helminths [37] . Other factors , such as drinking water sources , could play decisive role in the occurrence of these parasites . Therefore , water sources were explored in this study . Negative ratio values between high prevalence of E . histolytica/dispar , E . coli , and E . nana and use of tap water ( OR less than 1 ) show that contamination with these parasites decreases when use of tap water increases . Tap water usually undergoes chemical treatment to remove a number of infectious agents before being distributed to people . These precautions provide relatively good water quality . . Its consumption contributes to the reduction of infection by protozoa . Positive odds ratio ( OR greater than 1 ) , obtained between high rates of infection by G . lamblia , E . coli , and E . nana species and use of ponds , backwaters , and rivers as sources of household or drinking water , show that the prevalence of these parasites increase when the use of these sources increases . In Agboville , as in majority of developing countries , hygiene conditions are poor and could support propagation of G . lamblia through pond water contamination by human feces . In addition , animals such as rats bathe or drink in ponds and then leave many Giardia cysts [38] . These water sources are usually highly polluted , especially in rainy seasons [39] , contaminated by rain runoff charged with parasite cysts from animals and human droppings . Consumption of these exposed waters , in an area with high rainfall like Agboville , would be the basis for population-wide parasite infection . As in Offompo village , which has few or no toilets at all [19] , other localities studied lack toilets . In villages , when toilets exist , they are not often used and people defecate in the open . This observation was made during a study conducted in a village in Senegal , where 24% of the subjects defecated in the open , despite the existence of toilets [40] . This behavior in the population favors the spread of protozoal cysts . In order to limit the development of resistant strains of pathogenic intestinal protozoa , some authors recommend focusing preventive efforts and to target chemotherapy [22] , [23] . For parasitosis control , spatial distribution is important [41] . In the Agboville area , chemotherapy treatment should target the most infested populations . Sanitation education of the population , especially on the risks of surface water use and precautions to be taken , must accompany this treatment . The importance of providing communities with safe drinking water should also be impressed upon communities and authorities . Implemented together , targeted chemotherapy and provision of safe drinking water will allow better control of these parasites in the study area . Intestinal protozoa are common in the Agboville area of Côte d'Ivoire with a high prevalence of the pathogenic species E . histolytica/dispar and G . lamblia . Polyparasitism was highly prevalent in this area . No major focus zone was observed in the spatial distribution of both species . This result shows that control or eradication efforts against these intestinal protozoa must take into account the whole area , with urgent chemotherapy treatments delivered to the most-infected population segments . A significant negative association was observed between infection with E . histolytica/dispar and household use of tap water . G . lamblia was significantly associated with household use of pond water . Parasites prevalence decreases when tap water is used and increases when surface water is used . This work will help to make populations and political powers aware of the importance of these parasites and the need for safe drinking water in all the localities of this area . It can also contribute to develop an integrated control program against these parasites in this area of Côte d'Ivoire , including prophylactic and chemotherapy measures . | According to WHO , intestinal amoebiasis caused by Entamoeba histolytica is the third principal parasitic disease responsible for mortality in the world . This protozoal parasite infects approximately 180 million individuals throughout the world , among whom 40 to 110 thousand die from it each year . Giardiasis , caused by another protozoan parasite , Giardia lamblia , infects approximately 200 million individuals throughout the world , is a frequent cause of diarrhea in children , and can have negative impact on growth and development . Unfortunately , these intestinal protozoa are taken into account in few epidemiologic studies . The investigation we carried out to determine prevalence and spatial distribution of these infections shows the importance of these parasites in the Agboville department in southeast Cote d'Ivoire . Determination of spatial distribution of these parasites will help to focus delivery of chemotherapy in this area . In addition , our description of the relation of sources of drinking water with these parasitic infections will contribute to the development of an integrated treatment program for these parasites in this area of Côte d'Ivoire . This work will help make the population and political powers aware of the importance of these parasites and the need for safe drinking water in all localities of this area . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"infectious",
"diseases/gastrointestinal",
"infections",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2010 | Prevalence and Spatial Distribution of Entamoeba histolytica/dispar and Giardia lamblia among Schoolchildren in Agboville Area (Côte d'Ivoire) |
Crimean-Congo Haemorrhagic fever Virus ( CCHFV ) is a rapidly emerging vector-borne pathogen and the cause of a virulent haemorrhagic fever affecting large parts of Europe , Africa , the Middle East and Asia . An isothermal recombinase polymerase amplification ( RPA ) assay was successfully developed for molecular detection of CCHFV . The assay showed rapid ( under 10 minutes ) detection of viral extracts/synthetic virus RNA of all 7 S-segment clades of CCHFV , with high target specificity . The assay was shown to tolerate the presence of inhibitors in crude preparations of mock field samples , indicating that this assay may be suitable for use in the field with minimal sample preparation . The CCHFV RPA was successfully used to screen and detect CCHFV positives from a panel of clinical samples from Tajikistan . The assay is a rapid , isothermal , simple-to-perform molecular diagnostic , which can be performed on a light , portable real-time detection device . It is ideally placed therefore for use as a field-diagnostic or in-low resource laboratories , for monitoring of CCHF outbreaks at the point-of-need , such as in remote rural regions in affected countries .
CCHFV is an RNA virus classified in the Orthonairovirus genus , of the family Nairoviridae , within the order Bunyavirales ( International Committee on Taxonomy of Viruses 2016 ) . It displays a high degree of sequence variability [1] with the S-segments of CCHF viruses being phylogenetically grouped into 7 clades ( Europe 1 , Europe 2 , Africa 1 , Africa 2 , Africa 3 , Asia 1 and Asia 2 ) [2][3][4] . CCHF is easily transmitted by close contact and causes a virulent haemorrhagic fever in humans for which there is no effective prophylaxis or treatment: in consequence it is classified by the Advisory Committee on Dangerous Pathogens ( ACDP ) as a Hazard Group 4 pathogen mandating maximum microbiological containment ( containment level 4; CL4 ) . CCHFV has been detected in large parts of the globe including Southern and Eastern Europe , Central Asia , Western China , the Middle East and most of Africa [1] [5] [6] . CCHF has been described as “one of the most rapidly emerging viral haemorrhagic fevers in Africa , Asia and Eastern Europe , ” by the WHO , with case numbers increasing in many countries in recent years [7] [8] [9] . The case fatality for CCHF is typically between 5–50% , but has been documented as high as 73% [1] [10] [11] . Ticks of the genus Hyalomma function as vector as well as natural reservoir of CCHFV [12] . The Hyalomma tick feeds on various vertebrate hosts and as a consequence CCHFV is carried by wild animals and livestock [13] [14] and can be transmitted to humans both by tick bite and by contact with infected bodily fluids [1] . With outbreaks occurring in rural areas; commonly in low-resource regions with limited access to conventional laboratory facilities , there is an urgent need for a simple , fast , reliable and portable diagnostic test [8] . This would enable rapid diagnosis and public health management of suspect human cases , as well as surveillance of the virus in vertebrate and tick populations in isolated locations . CCHFV infection has a propensity for nosocomial transmission , especially during the early stages of disease when symptoms are poorly recognised and laboratory diagnosis is not commonly requested or performed . In these circumstances it can lead to outbreaks of highly pathogenic disease sustained by human-to-human transmission [8] [10] [12] . Rapid detection of positive cases could lead to more timely and appropriate support for patients , including their isolation to prevent transmission and the protection of health care providers by initiation of barrier nursing techniques . More convenient tools for CCHFV surveillance in the environment would also facilitate our understanding of the natural fluxes of virus in populations and help develop effective countermeasures and timely interventions [8] . There has been a recent proliferation of research into next-generation molecular diagnostics with improvements in performance relative to traditional PCR [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] . Many of these adopt a single ( isothermal ) incubation temperature and a variety of non-thermal methods to separate duplex DNA , allowing amplification of a target region without the need for thermal cycling . Recombinase polymerase amplification ( RPA ) is a well-established isothermal molecular technique and compares favourably with other isothermal methods such as LAMP ( Loop-mediated isothermal amplification ) , with a rapid turnaround time and simple set-up . It is performed at a single low temperature ( 37–42 degrees ) , using a recombinase enzyme to separate the DNA duplex and single-stranded DNA-binding proteins to stabilise the open complex [16] , allowing amplification and detection with standard probe chemistries . As there is no thermal cycling , there is no time-constraint on the amplification as there is with PCR and amplification occurs continually . This makes the RPA method significantly faster , with amplification occurring within 3–5 minutes for high copy number samples [15] [16] [27] . The low RPA incubation temperature and high speed makes assay systems based on the RPA technology particularly amenable to field-use due to the low power requirements . This allows detection on simple portable devices , which can be small and lightweight [15] [16] , including both miniaturised isothermal real-time detectors and fully automated rapid point-of care devices [15] [16] . RPA and other isothermal methods have shown high tolerance to inhibitors present in crude preparations of patient samples and arthropod vectors [28] [29] [30] . The elimination of laboratory intensive extraction procedures simplifies and speeds up the assay set-up and streamlines the automated point-of-care device , potentially making it lighter and cheaper to both purchase and run . The aim of this work was to develop an RPA assay as an alternative to the existing RT-PCR methods [31] and provide a fast and fieldable diagnostic , which could be used to test for CCHFV with minimal sample preparation , allowing surveillance and public health management decisions in isolated regions . The RPA shows a high degree of flexibility , so it could also be used in a clinic setting or a traditional laboratory , providing very rapid turnaround of results ( <20 minutes ) , with the potential for high-throughput in a 96-well plate format . The aims of the study were successfully achieved , with the development of a rapid RPA-based isothermal diagnostic capable of detecting all 7 S-segment phylogenetic clades of CCHF within 10 minutes .
Sera samples tested in this study were previously taken for the laboratory diagnosis of CCHFV and were classified as clinical specimens . No samples were collected specifically for this work , thus ethical approval for the study design was not required . Samples were anonymised within Tajikistan , so investigators were only supplied with sequentially numbered samples . Sera samples were collected and stored within the Tajik Research Institute of Preventive Medicine , Tajikistan . Samples for testing at Public Health England , UK were sent in accordance with national guidelines for both Tajikistan and the UK . The CCHF RPA assay was designed using alignments of the S-segments of CCHF strains and examination for regions of genomic stability , together with an NCBI BLAST search to check for CCHF specificity . The aim was to find a region which would allow uniform detection of all of the 7 clades of CCHF , but would not detect any non-CCHF sequences . A series of primers and probes were designed within this region and a small selection of synthetic RNA fragments ( approximately 1 . 7kb ) were designed to enable primer-probe testing . The primer and probe sets were sequentially tested for detection of the synthetic templates . The lead primer and probe combination was taken forward for further validation . Primers were prepared by Integrated DNA Technologies ( IDT ) and an RPA EXO probe was synthesised by ATD BIO; all as HPLC purified material . Primer and probe stocks were prepared at 100μM in a Tris-EDTA buffer and diluted to 10μM in molecular grade dH20 . A primer mix was prepared to 5μM ( of both forward and reverse primers ) and both primer mix and probe stocks were frozen at -20°C in single use aliquots . Synthetic CCHF S-segment DNA fragments from a selection of Europe group 1 and 2 strains and Africa 1 and 3 strains were prepared by IDT ( AY277672 , position 1–1673 , DQ211638 , position 1–1659 , DQ211643 , position 1–1671 , NC005302 , position 1–1672 and U88411 , position 1–1686 ) with the addition of T7 and SP6 promoters at the 5’ and 3’ ends respectively ( see S1 Fig ) . RNA templates were prepared from the synthetic DNA using a T7 High Yield RNA synthesis kit ( NEB ) . Approximately 1μg of DNA was added per reaction to a 0 . 2ml PCR tube with 2μl each of 100mM ATP , GTP , UTP and CTP , 2μl RNA polymerase mix , 2μl 10X reaction buffer and sufficient molecular grade dH20 to make the reaction up to 20μl . The reaction was incubated at 37 °C for two hours in a thermocycler . The RNA template was then DNase-treated to remove the original template contamination; 70μl nuclease free dH2O was added per tube with 10μl 10X DNase I buffer and 2μl RNase-free DNase I ( NEB ) . The tubes were mixed and incubated for 15 mins at 37°C . The RNA was purified using a Qiagen RNeasy minikit and quantified using a Qubit broad range RNA kit ( Thermo-Fisher Scientific ) . Crude samples included human serum male AB ( Sigma H422-20ML ) , female Ixodes ricinus ticks ( Charles River ) and Surine standard -ve control urine ( Sigma S-020-50ML ) . Sample preparation included aliquoting of the human male serum and urine standard into single use aliquots and storage at -20°C and fridge temperature respectively . The tick samples were prepared as tick pools ( 10 ticks ) and frozen at -20 °C . The ticks were prepared by adding 300μl molecular grade water , transferring to Precellys-R tubes and homogenising using a Precellys tissue homogeniser ( 3X 20 seconds , with 30 second breaks ) . The homogenate was centrifuged for 5 minutes at 5900xg and the supernatant retained . A serial dilution of each of the neat samples was prepared by diluting in molecular grade water . The CCHF RPA assay was performed in a 50μl volume using a TwistAmp Exo-RT kit ( TwistDx Cambridge UK ) . A master mix was prepared , composed of the following/reaction; 4 . 2μl of a 5μM primer mix ( forward and reverse primer ) , 0 . 6μl of the 10μM Exo-probe , 29 . 5μl rehydration buffer and sufficient distilled water to make the reaction up to 50μl after addition of all assay components . Where crude samples were used , 20 units ( 0 . 5μl ) of an RNase inhibitor was also included ( RNaseOUT 40U/μl Invitrogen ) . The master mix was distributed into the wells of a 96-well PCR plate . 1–5μl of template ( together with 5μl crude sample if used ) was added and the reaction mixture combined with the lyophilised enzyme pellet , before returning to the plate . The supplied magnesium acetate was diluted to 140mM with molecular grade dH20 , 5μl was added to each well and the plate briefly centrifuged before running at 40 °C for 40 minutes on an Applied Biosystems 7500 real-time PCR system , with fluorescence detection every 60 seconds in the FAM channel . The threshold was set at 50 , 000 delta Rn . The RT-RPA basic assay was performed in a 50μl volume using a TwistAmp Basic-RT kit ( TwistDx Cambridge UK ) . A master mix was prepared , composed of the following/reaction; 4 . 2μl of 5μM primer mix ( forward and reverse primer ) , 29 . 5μl rehydration buffer and sufficient distilled water to make the reaction up to 50μl after addition of all assay components . The master mix was distributed into 0 . 2ml PCR tube strips . 5μl of template was added and the reaction mixture combined with the lyophilised enzyme pellet , before returning to the wells . The supplied magnesium acetate was diluted to 140mM with molecular grade dH20 and 5μl was added to each well . The tube strip was then briefly centrifuged before incubating at 40 °C for 40 minutes on a thermocycler . The products of the RPA were purified using a Qiagen QIAquick PCR purification kit , then run on an Invitrogen 2% Agarose gel ( E-Gel EX with Sybr Gold II ) with an Invitrogen E gel 1Kb plus ladder . This method was adapted from the paper by Atkinson et al 2012 [31] . Briefly , the CCHF RT-PCR assay was performed in a 20μl volume using a Superscript III Platinum One-step quantitative RT-PCR kit ( Invitrogen ) . For each reaction , a 15μl master mix was prepared containing 10μl of the supplied 2X reaction mix , 0 . 8μl of superscript III RT/Taq enzyme mix and final concentrations of each primer of 1 . 2μM and probe of 0 . 8μM . Where crude samples were used , 20 units ( 0 . 5μl ) of an RNase inhibitor was also included ( RNaseOUT 40U/μl Invitrogen ) and sufficient molecular grade water was added for a final reaction volume of 20μl , once all assay components were included . The master mix was distributed into the wells of a 96-well PCR plate . 3–5μl of template ( together with 5μl crude sample if used ) was added and the plate briefly centrifuged before being run on an Applied Biosystems 7500 real-time PCR system . The PCR run parameters included a 10 minute RT step at 50 °C , followed by a 2 minute denaturation step at 95°C , an amplification stage composed of 45 cycles of denaturation: 95°C for 10 seconds , and annealing/extension at 60°C for 40 seconds , followed by a final extension at 40 °C for 20 seconds . Fluorescence was detected in the FAM channel , once each cycle during the amplification stage . The threshold was set at 250 , 000 delta Rn . A collection of CCHFV strains representing each of the following S-segment clades: Asia 1 , Asia 2 , Africa 2 , Africa 3 and Europe 1 were cultured and viral RNA was extracted using a standard RNA extraction kit ( QIAamp viral RNA kit ) . Synthetic whole S-segment viral RNA was used to represent Africa 1 and Europe 2 clades . Viral RNA extracts covering Mammarenavirus , Marburgvirus , Henipavirus , Flavivirus , Alphavirus and the Orthohantavirus genera were donated by the Rare and Imported Pathogens Laboratory , PHE Porton from a collection of standard diagnostic assay positive controls . They are described as positive , with a Ct of approximately 30 in their respective real-time assays , or as having a clear and distinct band after electrophoresis of PCR products following a conventional block-based PCR . The Orthonairovirus samples Hazara and Issyk-Kul were supplied in-house by the National Collection of Pathogenic Viruses and the Virology and Pathogenesis group respectively and were confirmed positive using a block-based PCR . Ticks identified as Hyalomma anatolicum were collected from the Gisar , Hissor , Kulob , Rumi , Jillikul and Hamadoni districts of Tajikistan and prepared as pools of 10 ticks by the Tajik Research Institute of Preventive Medicine , Dushanbe , Tajikistan . The ticks were added to 1ml Qiagen AVL buffer and were homogenised using a disposable plastic pestle and mortar , allowing 10 minutes post-homogenisation for inactivation . Sera samples were collected from suspected CCHF patients from the Rudaki , Nahiyeh Voseh , Bokhtar , Kologh , Nahiyey farkhor , Moominabad , Khatlon , Bolgevon , Tursonzadeh , Kabadian , Dangarah , Vosse and Kulob districts of Tajikistan by the Tajik Research Institute of Preventive Medicine , Dushanbe , Tajikistan . The sera samples were heat-inactivated for 30 minutes at 56°C and both the inactivated tick pool samples and sera were frozen at -80 for storage . The tick homogenate was centrifuged at 4000xg for 10 minutes to pellet the particulate matter . A 140μl volume of the neat sera samples and tick pool supernatant were extracted using a Qiagen QIAamp viral RNA mini kit , with elution into 80μl elution buffer . Note that most sera samples represent a single sera collection , taken on a single day from separate patients . Where samples are multiple sample collections , taken on separate days from a single patient , the sample number is followed by a letter ( e . g . sample 4a , 4b , 4c ) and a collection number is given . Genbank accession numbers/ NCBI reference sequences for genes used in this study: Bagdad 12 ( Genbank accession number AJ538196 ) , Dubai 616 ( Genbank accession number JN108025 ) , DAK 8194 ( Genbank accession number U88411 ) , Semunya ( Genbank accession number DQ076413 ) , Congo 3010 ( Genbank accession number DQ144418 ) , SPU4/81 ( Genbank accession number DQ076416 ) , IbAr10200 ( NCBI reference sequence NC_005302 ) , Kosovo Hoti ( Genbank accession number DQ133507 ) , ROS/T128044 ( Genbank accession number AY277672 ) , Drosdov ( Genbank accession number DQ211643 ) , AP92 ( Genbank accession number DQ211638 ) .
The CCHFV RPA assay was developed by making alignments of CCHF S-segment sequences and analysing them for stable regions to identify sequences that could be used as the basis for primer and probe design . The lead primer-probe set ( see Fig 1A ) is shown aligned to a selection of CCHFV strains representing the 7 S-segment clades of the virus ( Africa 1 , Africa 2 , Africa 3 , Asia 1 , Asia 2 , Europe 1 and Europe 2 ( Fig 1B ) . The regions chosen for the RPA primer design were in the vicinity of the primer binding sites of our in-house CCHF PCR , which has shown good cross-clade detection [31] . Some single base mismatches within the primer and probe regions remained , but this was unavoidable due to the high intrinsic sequence variability of the CCHFV S-segment . We were confident that our RPA assay would show good amplification across the clades however , as there was significant overlap in the design region with the PCR and the RPA primers were generally well matched and long , thereby increasing resilience to minimal base-mismatches . The target region was also subjected to a BLAST ( blast . ncbi . nlm . nih . gov ) search to check for cross-reactivity with non-CCHFV sequences; no non-CCHFV sequences were identified , suggesting that the assay would be CCHFV-specific . The detection limit of the CCHFV RPA assay was determined by testing with a serial dilution of a synthetic RNA S-Segment of the Europe 1 strain AY277672 from 5X106 template copies down to 50 copies ( Fig 2 ) . Rapid detection was observed , occurring within 5 minutes for the detection of 50X106 template copies , just over 10 minutes for the detection of 5X103 copies and 16 minutes for the detection of 500 copies . The detection limit was between 500 and 50 copies , with strong and rapid detection down to 500 copies , but with variable and sub-threshold detection at 50 copies . The probit analysis gave a predictive value for the limit of detection of this assay as 251 copies of target . The data showed a good correlation between copy number and time to positive , suggesting that this method is either quantitative or semi-quantitative . The CCHFV RPA was subsequently tested against a selection of strains representing all 7 extant clades of the viral S-segment . This included where possible cultured viral extract , and a selection of 5X105 copies/reaction synthetic whole S-segment RNA templates ( See S1 Fig . for synthetic target sequence information ) . The target was successfully identified ( Fig 3 ) , across all 7 clades , with the same rapid detection noted earlier . The RPA time to positive ( TTP ) was less than 12 minutes for all viral extracts , with most detected between 5–10 minutes . Data from a CCHFV S-segment RT-PCR [31] is shown alongside , with the TTP values ranging from 39 . 6 to 54 . 6 minutes . The CCHFV RPA was also performed with a basic RT-RPA kit ( no exo-probe ) and synthetic CCHFV RNA templates . When analysed by gel electrophoresis with a DNA intercalating dye , the results showed a single band of the expected size of around 150bp , suggesting that only the target region is amplified ( see S2 Fig ) . The specificity of the CCHFV RPA was tested against a panel of viral extracts , encompassing both phylogenetically related strains and species which cause a similar haemorrhagic aetiology in humans and may be tested as part of a differential diagnosis . These included viruses from the Mammarenavirus , Marburgvirus , Henipavirus , Orthonairovirus and Orthohantavirus genera ( Fig 4 ) . All produced negative results , backing up the hypothesised specificity which had been suggested from the bioinformatic information . Of particular note is the fact that the closely-related Orthonairoviruses Hazara and Issyk-Kul were negative . Isothermal molecular amplification methods such as RPA and LAMP are tolerant to crude samples , allowing detection in patient bodily fluids , arthropod preparations and other crude material with minimal processing and avoiding the need for extraction [28] [29] [32] [33] [34] . The inhibitory effect of interfering agents within a crude sample on the CCHFV RPA was studied by spiking a known quantity of synthetic CCHFV template into crude preparations of human serum , a homogenised tick pool and synthetic human urine . Fig 5A shows strong detection in the presence of each crude sample . There is however evidence of some inhibition at the higher crude sample dilutions , with higher TTP values , although the urine sample showed no observable inhibition at any concentration . Serum showed full detection of the 5X106 copies/reaction template at the 1-in-10 dilution and the tick preparation at the 1-in-100 dilution . This suggests the CCHFV RPA can tolerate the inhibitors present in crude samples with minimal sample preparation , when the copy number of target is high . We next examined the limit of detection in minimally diluted crude samples . Fig 5B indicates that the urine and tick preparation had no effect on the assay limit of detection , with both showing full detection down to 500 copies and partial detection at 50 copies , with comparable TTP values to data using un-spiked purified RNA . Serum showed a minimal effect on the detection limit , with partial detection at 50 and 500 copies and higher TTP values , but with comparable TTP and 100% detection at 5000 copies . Overall however this work suggests that the inhibitory effect of diluted crude samples on the CCHFV RPA is minimal . All crude sample preparations except for the urine showed some inhibitory effect on the detection limit of the CCHFV RT-PCR , with approximately a 10-fold change for the tick and serum preparations . CCHFV has caused seasonal outbreaks [35] of disease in Tajikistan since its discovery in the region over 40 years ago . The assay including a portable incubator and reader ( Optigene Genie III ) was taken to a collaborating laboratory in Dushanbe; the Tajik Institute for Preventive Medicine . This Ministry of Health institute undertakes diagnostic testing for a range of serious pathogens and is the National Centre for CCHF diagnosis; it has regular access to CCHF clinical samples and supports national disease reporting programmes . The CCHFV RPA was set up and used on a collection of extracted patient sera samples and environmental tick extracts obtained in relation to outbreaks of CCHF in 2013–2015 in Tajikistan , these field samples were tested alongside a standard RT-PCR assay [31] that had already been established . The field samples were then sent back to our UK-based laboratory for further testing . The CCHFV RPA ( Fig 6 ) detected nearly all confirmed positives ( 88% of 8 positive tick samples and 100% of 13 positive sera samples ) , missing only one very late TTP weakly positive tick extract . The RPA also verified the negatives ( 100% of 8 negative tick samples and 91% of 11 independent ( from separate patients ) negative sera samples ) , with one exception ( sample 16 ) . It is possible this could be a false positive , however this was a sample which had previously been tested by ELISA and found to be IgM positive for CCHF , so it is likely that this is actually a positive sample . The majority of positive samples were detected in less than 20 minutes ( 4 samples were detected between 21 and 31 minutes ) .
A CCHFV RPA has been successfully developed and shown to rapidly detect all 7 CCHFV S-segment types , despite the presence of a small number of single mismatches within the primer and probe binding sites . Data from a negative panel of closely related viruses including those that might be included as part of a differential diagnosis , illustrate the high specificity for CCHFV . The sensitivity of this assay was statistically estimated to be 251copies of RNA target , which compares well with other published RPA assays [20] [21] [23] [24] . While this sensitivity is lower than the published RT-PCR for CCHFV of 5–50 copies [31] , it is still within the clinical detection range present in most acute CCHF patients , as demonstrated by its ability to pick-up all of the confirmed sera positives in the panel of samples from an outbreak in Tajikistan . The assay is also very rapid , providing clear results within 20 minutes ( 5 minutes for the detection of 50X106 template copies , 10 minutes for the detection of 5X103 copies and 16 minutes for the detection of 500 copies ) and an average of 13 . 8 mins ( range 6 . 8–30 . 7 mins ) for the real clinical/field samples . The CCHFV RPA performed well with spiked crude samples with only a small level of dilution required to remove inhibitory effects . In addition the CCHF RPA strongly detected a clinical sample which had been missed by the PCR , suggesting that the RPA may be able to detect a wider cross-section of CCHF strains than the existing PCR . This is supported by our cross-template detection study , in which the RPA detected a strain Drosdov , which the PCR failed to pick up . The CCHF RPA has demonstrated its potential for use both as a field/low resource laboratory diagnostic and traditional laboratory test . It is significantly faster than existing RT-PCR-based methods ( maximum run time of 35 minutes , compared to 1 hour 10 minutes of the RT-PCR ) enabling faster-turnaround and key information to physicians and health care workers , underlining its suitability as a point-of-need diagnostic in the community clinic , or the field . The fast turnaround would also be of benefit in a standard laboratory set-up enabling high-throughput testing . The potential of this diagnostic in low resource settings is highlighted by its validation in Tajikistan using a simple portable isothermal real-time detector . This mirrors published literature generated with the Genie III device [36] and underlines the benefits of an isothermal diagnostic over the traditional RT-PCR in low resource settings . The applicability of the RPA and other isothermal molecular assays to crude unprocessed clinical samples is often highlighted [28] [29] [30] as a useful advantage . While we show that the current assay is amenable to such samples , further studies will be needed to fully evaluate the potential for use of this assay with crude clinical material , including careful consideration of the biosafety implications when handling this HG4 virus . This will involve testing the RPA with live virus preparations spiked into crude sample material in addition to un-extracted virus in real clinical samples , to confirm whether native virus can be detected . The primary consideration when working with infectious viral material will be to examine methods for disrupting the viral envelope to allow access to the viral nucleic acid and to inactivate the virus to render it safe to handle , whilst avoiding adding to the complexity of the process . | The Crimean-Congo Haemorrhagic fever ( CCHF ) recombinase polymerase amplification assay is a new , rapid and portable diagnostic method . It has been developed for the detection of infection with CCHF virus ( CCHFV ) , the cause of a deadly haemorrhagic disease in humans and an emerging global health threat . As a rapid diagnostic suitable for use on a portable and lightweight detection device , this has the potential to be used for fast-turnaround diagnosis at the point of need , providing timely results to clinicians at the bedside and preventing the spread of virus in the hospital setting . As a portable diagnostic it would also enable the diagnosis of CCHFV to be taken out of the laboratory during an outbreak , enabling testing in field laboratories or community clinics . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"reverse",
"transcriptase-polymerase",
"chain",
"reaction",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"crimean-congo",
"hemorrhagic",
"fever",
"tropical",
"diseases",
"rna",
"extraction",
"geographical",
"locations",
"urine",
"neglected",
"tropical",
"dis... | 2017 | A recombinase polymerase amplification assay for rapid detection of Crimean-Congo Haemorrhagic fever Virus infection |
Dengue virus is endemic globally , throughout tropical and sub-tropical regions . While the number of epidemics due to the four DENV serotypes is pronounced in East Africa , the total number of cases reported in Africa ( 16 million infections ) remained at low levels compared to Asia ( 70 million infections ) . The French Armed forces Health Service provides epidemiological surveillance support in the Republic of Djibouti through the Bouffard Military hospital . Between 2011 and 2014 , clinical and biological data of suspected dengue syndromes were collected at the Bouffard Military hospital and analyzed to improve Dengue clinical diagnosis and evaluate its circulation in East Africa . Examining samples from patients that presented one or more Dengue-like symptoms the study evidenced 128 Dengue cases among 354 suspected cases ( 36 . 2% of the non-malarial Dengue-like syndromes ) . It also demonstrated the circulation of serotypes 1 and 2 and reports the first epidemic of serotype 3 infections in Djibouti which was found in all of the hospitalized patients in this study . Based on these results we have determined that screening for Malaria and the presence of the arthralgia , gastro-intestinal symptoms and lymphopenia < 1 , 000cell/ mm3 allows for negative predictive value and specificity of diagnosis in isolated areas superior to 80% up to day 6 . This study also provides evidence for an epidemic of Dengue virus serotype 3 previously not detected in Djibouti .
With 3 , 900 million people exposed in 128 countries , the Dengue virus ( DENV ) is the most common arboviral disease in the world . Every year , 50 to 100 million people are infected and at least 30 , 000 people die [1–3] . Four antigenically distinct serotypes ( DENV1-4 ) circulate simultaneously in endemic countries [4] . The virus is endemic globally in tropical and sub-tropical regions . Primarily in Asia , South and Central America [5] , however , reports on Dengue circulation are increasing in Africa [6 , 7] . Although the number of epidemics due to the four serotypes is markedly increased in East Africa [8] and to a lesser extent in West Africa [9] , the number of cases reported in Africa ( 16 million infections ) remain at very low level compared to Asia ( 67 million infections ) and nearly equivalent to that of the Americas ( 13 million infections ) [10] . The first epidemic of DENV in Djibouti was reported in 1991–1992 with 12 , 000 cases of serotype 2 [11] . Introduction of DENV1 in Djibouti was detected in 1998 with sporadic cases until 2000 [7] . Between 2000 and 2002 , 185 cases of Dengue virus were detected by serology with only 6 identified as serotype 1 by virus isolation [12] . Due to the continuous geographic expansion of Dengue disease , the World Health Organization implemented a surveillance program [3] . The French Armed Forces Health Service ( FAFHS ) participates in this program through its overseas sentinel network [13] . In the Republic of Djibouti , the public health system relies on the activity of the medical and surgical hospital Bouffard ( HMC Bouffard ) . Due to its diversified technical equipment , HMC Bouffard cares for the overseas community and the Djiboutian people This study provides an analysis of epidemiological , clinical and biological data related to suspected and confirmed cases of Dengue managed at HMC Bouffard between 2011 and 2014 . Based on the observed symptoms , the complete blood count data , and the biological diagnosis , this study aims to help practitioners working in isolated areas to establish Dengue etiology in resource limited settings .
This retrospective study presents the clinical and biological analysis results of Dengue-suspected cases managed at HMC Bouffard between January 2011 and May 2014 . Dengue-like cases were defined when patients showed one or more of the following signs: fever , arthralgia , headache , myalgia , gastro-intestinal symptoms , hemorrhagic syndrome and visceral organ failure . Every patient was tested for Malaria using the HRP-2 test ( Core Malaria , Core Diagnostics , Birmingham , UK ) and the QBC method . In this study , we considered as Dengue-suspected cases patients presenting a Dengue-like syndrome and testing negative for malaria . These patients were recruited through the emergency department of the hospital ( admissions to the hospital for medical checkups due to fever represent 6% of the patients ) or directly at the medical laboratory when they came with a medical prescription issued by civilian or military physicians practicing in Djibouti . A case of dengue was confirmed ( DENV-POS ) when a suspected patient test produced at least one positive result for one of the following diagnostic tests: NS1 Antigen ( NS1 Ag ) , anti-Dengue Immunoglobulin M ( IgM ) or real-time polymerase chain reaction ( RT-PCR ) . Isolation of Flavivirus specific IgG alone is insufficient for a diagnosis of recent DENV infection . Suspected patients who had negative results for the three DENV diagnostics were considered as negatives ( DENV-NEG ) . For each suspected case , an epidemiological form ( C11 syndrome « dengue-like » ) provided by the FAFHS was filled in to gather demographic , clinical and epidemiological data . Biological sampling was done in order to achieve a complete blood count and the confirmatory diagnosis of Dengue cases . The following data come from the C11 forms: number of Dengue cases among the suspected cases , periodicity , serotype , clinical signs , platelet and lymphocyte counts at the physician visit . Diagnosis for arboviruses for all the French military hospitals , and for the French soldiers in Africa is assessed by the French National Reference Center ( NRC ) for Arboviruses belonging to the French Armed Forces Biomedical Research Institute . As the Reference Center , arbovirus diagnosis was realized only when all the required information was provided: clinical and biological data , travel history and date of onset of symptoms . All the data presented in this paper were used for the routine diagnosis . For each sample , a complete blood count was automatically performed ( Pentra 120 , ABX company ) at the Bouffard hospital laboratory . No follow-up of the platelet and lymphocyte counts were performed . All the samples collected in the range of 0 to 7 days after onset of symptoms were tested in order to detect DENV viral RNA and viral NS1 Ag . The NS1 Ag was tested with the SD BIOLINE Dengue Duo rapid test ( NS1 Ag + IgG/IgM ) ( Standard Diagnostics , Inc , Republic Of Korea ) according to the manufacturer's instructions . It is a rapid , in vitro immunochromatographic , one-step assay designed to detect both DENV NS1 antigen and IgG/IgM antibodies to DENV in human serum , plasma or whole blood . Viral RNA was extracted from 140μl of serum using QIAamp Viral RNA Mini Kits ( Qiagen , Germany ) by the French NRC for arboviruses . Five systems of real-time RT-PCR assays were used to detect all viral strains and identify the 4 serotypes , as previously described by Leparc-Goffart et al . [14] . Each sample collected at the HMC Bouffard was analyzed with the rapid test SD BIOLINE Dengue Duo ( NS1 Ag + IgG/IgM ) . Concurrently , each sample collected at least 5 days after the onset of symptoms was serologically analyzed ( IgM and IgG ) to detect DENV , Chikungunya , West Nile and Rift Valley Fever infections . In-house IgM antibody capture enzyme-linked immunosorbent assay ( MAC-ELISA ) and IgG indirect ELISA were used to detect IgM and IgG antibodies to DENV [15] . After a first step focused on the descriptive analysis of the study population , platelet and lymphocyte counts were compared as a function of Dengue status ( DENV-POS versus DENV-NEG ) and then among DENV serotypes in DENV-POS patients . The statistical tests used were non-parametric tests , as conditions for application of parametric tests were not fulfilled . A Mann-Whitney test was used to compare the median counts as a function of the dengue status and a Kruskal-Wallis test was used to compare the median counts as a function of the dengue serotype . Then , a univariate analysis was performed to determine the relationship between the different symptoms or clinical signs and the Dengue diagnosis . The statistical tests used were the Chi-squared test , or the Fisher’s exact test when the conditions for application of the Chi-squared test were not fulfilled . A multivariate analysis was finally performed to determine the main independent symptoms or clinical signs associated with Dengue diagnosis . All the independent variables achieving a p-value ≤ 0 . 20 at the univariate stage were introduced in the multivariate analysis model . Odds ratios were obtained using logistic regression models . Finally , the same analysis ( univariate analysis followed by a multivariate analysis ) was performed for the DENV serotype among the DENV-POS patients , using multinomial regression models . The statistical significance was defined as a p-value <0 . 05 . All data were analyzed using the Stata Statistical Software v12 ( Statacorp , USA ) .
Between 2011 and 2014 , 354 Dengue-suspected cases , presenting with one or more Dengue-like symptoms , were cared for in HMC Bouffard ( Fig 1 ) . A total of 128 dengue cases , as well as one West Nile virus case , one Rift Valley fever case and one imported Zika virus case from French Polynesia were diagnosed and confirmed by the Arboviruses NRC . Most of the DENV-POS patients were male ( 66% ) . Military personnel were highly represented , comprising more than 30% of the dengue positive cases . The remaining 70% corresponded to the military personnel’s family , the expatriate population and some Djiboutian civilians . Twelve cases presented only a positive serology ( presence of IgM and anti-flavivirus IgG antibodies ) , 38 cases were only NS1-positive . A total of 78 dengue cases were diagnosed by PCR . Most of the cases detected by PCR ( 74/78 ) could be serotyped: there were 9 cases of DENV1 ( 12% ) , 25 cases of DENV2 ( 34% ) and 40 cases of DENV3 ( 54% ) ( Fig 2 ) . Cases of DENV1 occurred between November 2011 and June 2012 , with a recurrence in June 2014 . The epidemic of DENV3 began in December 2011 , reached an epidemic peak during the summer of 2012 and declined the winter of 2012–2013 . A few cases of DENV3 were reported during the spring of 2013 . Cases of DENV2 occurred during a short period , in epidemic form , from December 2013 to March 2014 , with 23 cases reported in 4 months . Clinical signs were studied from 98 DEN-POS files ( i . e . 77% of the positive cases ) . The other files were not taken into account as patients were addressed without clinical data to the hospital laboratory by a physician from the inner-city . Among the analyzed cases , 98% suffered from fever , 89% from arthralgia , 75% from headache , 39% from gastrointestinal symptoms such as nausea and vomiting , 35% from retro-orbital pain and 22% from a cutaneous rash . No atypia was identified . Ten patients were admitted to the hospital , i . e . 7 . 8% of the DENV-POS patients . Among them , 6 could be serotyped and were infected with the DENV-3 virus . In 6 cases , the criteria for hospitalization were a dehydration caused by digestive disorders and hyperthermia . Two female patients were admitted in the department of intensive care with severe Dengue: an 8-year old girl for a Dengue Hemorrhagic Fever , in a state of shock and with multiple organ failure and a 50-year old woman , without any comorbidity , for an isolated acute pulmonary edema , without any hemorrhagic sign . No deaths were reported . After univariate analysis according to Dengue status ( comparison of DENV-POS and DENV-NEG patients' clinical signs ) , the presence of arthralgia was the only factor significantly associated to Dengue disease ( p = 0 . 007 ) . This association persisted after multivariate analysis , as patients presenting arthralgia were 2 . 5 times more likely to be DENV-POS ( 95% confidence interval of adjusted OR: 1 . 2–5 . 2 ) , adjusted for the presence of gastrointestinal symptoms ( adjusted OR: 1 . 6; 95% CI: 0 . 9–2 . 7 ) ( Table 1 ) . Presence of retro-orbital pain and skin rash seemed to be more frequent for DENV3 cases ( Fig 3 ) . However , no statistically significant association between these symptoms and serotypes was found ( respectively p = 0 . 13 and p = 0 . 39 ) . The lymphocyte count at admission was significantly lower in DENV-POS patients than in DENV-NEG patients ( respectively 700 cells/mm3 vs 1 , 330 cells/mm3 ) ( p<0 . 001 ) ( Table 2 ) . The lowest lymphocyte count was measured in a DENV-POS patient being tested of two days after the onset of the symptoms at 200 cells/mm3 . In the patients being tested after the 5th day of clinical evolution , the lymphocyte count returned to basal levels . Finally , the median lymphocyte count at admission was significantly lower in serotype-3 than in serotype-1 and serotype-2 ( p-value = 0 . 02 ) ( Table 2 ) . Regarding platelet counts , it appears that despite median platelet count being significantly lower in DENV-POS patients than in DENV-NEG patients ( Table 2 ) , thrombocytopenia remained inconstant ( median platelet count of 159 G/L among DENV-POS patients ) and occurred mainly between the 3rd day and the 7th day post onset . Median platelet count did not vary according to DENV serotype ( p = 0 . 54 ) ( Table 2 ) . The negative predictive value ( NPV ) related to the combination of fever , arthralgia , lymphopenia and its specificity were analyzed . In our study , the NPV equaled 100% at Day 1 post onset of symptoms and was superior to 80% till Day 6 . The specificity increased from 40% at Day 1 to 80% at Day 6 post onset of symptoms . The NPV related to the combination of fever , arthralgia , gastro-intestinal symptoms and lymphopenia inferior to 1 , 000cells/mm3 from Day 1 to Day 7 post onset of symptoms was superior to 70% and its average specificity remained superior to 85% . Moreover , among the DENV-POS cases presenting the combination of fever , arthralgia , and gastro-intestinal symptoms , 8 . 3% tested without lymphopenia and 91 . 7% with lymphopenia . In contrast , we observed the same proportion of DENV-NEG cases presenting fever , arthralgia , gastro-intestinal symptoms without and with lymphopenia ( 47 . 8% and 52 . 2% respectively ) . Among the patients presenting the combination of fever , arthralgia , and gastro-intestinal symptoms without lymphopenia , 15 . 3% are DENV-POS and 84 . 7% are DENV-NEG . Among the patients presenting the combination of fever , arthralgia , and gastro-intestinal symptoms with lymphopenia , 64 . 7% are DENV-POS and 35 . 3% are DENV-NEG .
Between 2011 and 2014 , with 128 ( 36 . 2% ) cases diagnosed among 354 suspected cases , the rate of detected Dengue cases was greater than the one observed by Houze in 2003 ( 12 ) . In that retrospective study , the author described 185 cases ( confirmed by serologies and/or viral cultures ) between January 2000 and March 2003 found among 1 , 224 patients with a non-malarial fever . This increase in Dengue case declaration could be explained by the emergent epidemiological profile of Dengue , by a more exhaustive patient recruitment , but also by the improvement of biological diagnostic tests facilitating Dengue virus detection ( NS1 Ag detection at the Bouffard hospital laboratory and RT-PCR assays performed by the Arboviruses NRC ) , resulting in an increase of case notification . However , the incidence of Dengue is definitely underestimated , due to the recruitment bias associated with the administrative accessing rules to the Bouffard hospital and to the recurrent lack of rapid diagnosis test at the laboratory ( medical supply problems ) . Effectively , hospital consultation is mainly restricted to an affluent population , which is better protected against mosquito attacks and consults at the earliest stages of the disease . Finally , the study was restricted to symptomatic patients , with all asymptomatic patients ruled out . Dengue virus is symptomatic in only around 50% of cases [5] . With an epidemic of 12 , 000 cases reported by Rodier et al . [11] , active surveillance of Dengue began in Djibouti in the early 1990’s . Since then , it has been circulating in an endemo-epidemic pattern . Each year between December and May , an epidemic peak is reached , due to the abundance of its vector ( peak in the number of mosquitoes due to an increase in rainfall between December and May ) . In 2012 , following the heavy spring rainfalls , an increase in the number of cases was observed until June . For the first time in Djibouti , serologic typing of dengue cases that arose during the epidemic in winter 2011–2012 , confirmed DENV3 cases . An epidemic of DENV3 has been described in 2010 in Sudan [16] . Another epidemic occurred in Kenya in October 2011 , with about 5 , 000 suspected cases , in a border district with Somalia and Ethiopia [17] . DENV1 appears to circulate in an endemic pattern . During the surveillance period of our study , no DENV4 case was diagnosed , as expected this genotype is sparsely reported in Africa [18] . However , due to the lack of a national arboviral surveillance system , Dengue circulation in the Horn of Africa is difficult to assess . Our results highlight a higher rate of arthralgia in DENV-POS patients than in DENV-NEG patients . Despite no statistically significant association found in this study , patients presenting with gastrointestinal symptoms tended to be more likely to have Dengue . No statistical link was found between clinical signs and DENV serotypes . This could be due to a lack of sampling power . During the first week of symptoms evolution , analyzing data from the complete blood count demonstrated an early and almost constant lymphopenia in DENV-POS adult patients , which has already been described [19] . This study also demonstrated that a lymphocyte count below 1 , 000 cells/mm3 is associated with Dengue . It was already reported that lymphopenia is associated with severe forms of Dengue [20] . A lymphocyte count higher than 1 , 500 cells/mm3 was unlikely to be associated with the diagnosis of Dengue in our study . Even if platelet counts are lower in DENV-POS patients , thrombocytopenia is inconstant in DENV-POS patients on the day of diagnosis . This is due to the fact that thrombocytopenia is delayed in uncomplicated Dengue cases in adult patients [19] . Biological and clinical data collected during this study offer a direction in diagnosis for physicians working in isolated areas , suggesting that Dengue-suspected patients with arthralgia , gastrointestinal symptoms and a lymphopenia below 1 , 000cells/mm3 are very likely to be DENV-POS . A 10 day follow up of DENV-POS patients is recommended with paracetamol treatment until the cessation of the symptoms . Among 354 Dengue suspected cases , only 3 cases could be attributed to a known arbovirus other than Dengue leaving 63% of the non-malarial Dengue-like cases undetermined . No infection caused by Chikungunya virus was recorded during this period . However , Borgherini et al . [21] have shown that infection caused by Chikungunya virus is also associated with lymphopenia and severe thrombocytopenia . Nkoghe et al . [22] put forward that gastrointestinal symptoms are similar in Chikungunya and in Dengue infections and that in both cases lymphopenia below 1 , 000 cells/mm3 can be noted . Similar symptoms may also be found in infections due to other arboviruses , such as Zika infection for example . Data of the present study may therefore help guiding the diagnosis in favor of an arbovirus , if they are not in favor of Dengue . The 2 severe clinical cases observed during this study were associated with DENV3 . There is no clear data on the link between serotypes and severity of the infection in previous studies [23 , 24] but severity seems to depend on a previous exposure to another serotype ( Halstead's controversial theory about immunological facilitation ) [25] . Moreover , retro-orbital pain and skin rash are more frequent for DENV3 . Biologically , lymphopenia below 800 cells/mm3 also suggests the presence of DENV3 . However , these results were not observed in previous studies [23] and should therefore be confirmed on larger numbers of patients . After sequencing these strains , we have determined that they belonged to genotype III , often associated to more severe symptomatic forms [24 , 26] . Although Dengue has only recently come up in Africa [27] , we expect that DENV3 will spread throughout the continent in the coming years , with more severe Dengue forms in naive countries . | Dengue virus is emerging worldwide , however , little is known about the burden of Dengue in Africa . Effectively , only sporadic cases and few epidemics have been reported in the last 30 years . This descriptive study reports clinical and biological data of Dengue-suspected cases analyzed in the Bouffard military hospital of Djibouti from 2011–2014 . It also confirms the circulation of Dengue virus serotypes 1 and 2 and reports the first epidemic of Dengue virus serotype 3 infections in Djibouti . Directions for diagnosis are offered to practitioners working in resource limited settings and dealing with a Dengue-like syndrome of less than seven days and negative for malaria . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"lymphopenia",
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"geographical",
"locations",
"microbiology",
"viruses",
"signs",
... | 2016 | Clinical Survey of Dengue Virus Circulation in the Republic of Djibouti between 2011 and 2014 Identifies Serotype 3 Epidemic and Recommends Clinical Diagnosis Guidelines for Resource Limited Settings |
Rose-comb , a classical monogenic trait of chickens , is characterized by a drastically altered comb morphology compared to the single-combed wild-type . Here we show that Rose-comb is caused by a 7 . 4 Mb inversion on chromosome 7 and that a second Rose-comb allele arose by unequal crossing over between a Rose-comb and wild-type chromosome . The comb phenotype is caused by the relocalization of the MNR2 homeodomain protein gene leading to transient ectopic expression of MNR2 during comb development . We also provide a molecular explanation for the first example of epistatic interaction reported by Bateson and Punnett 104 years ago , namely that walnut-comb is caused by the combined effects of the Rose-comb and Pea-comb alleles . Transient ectopic expression of MNR2 and SOX5 ( causing the Pea-comb phenotype ) occurs in the same population of mesenchymal cells and with at least partially overlapping expression in individual cells in the comb primordium . Rose-comb has pleiotropic effects , as homozygosity in males has been associated with poor sperm motility . We postulate that this is caused by the disruption of the CCDC108 gene located at one of the inversion breakpoints . CCDC108 is a poorly characterized protein , but it contains a MSP ( major sperm protein ) domain and is expressed in testis . The study illustrates several characteristic features of the genetic diversity present in domestic animals , including the evolution of alleles by two or more consecutive mutations and the fact that structural changes have contributed to fast phenotypic evolution .
Rose-comb ( Figure 1 ) was one of the autosomal dominant traits William Bateson used in his seminal paper describing Mendelian inheritance in animals for the first time [1] . This mutation probably occurred early in the process of chicken domestication , as it is widespread among chicken populations originating in both Asia and Europe , separated for hundreds of years . A few years after the first description of the mode of inheritance of Rose-comb , Bateson and Punnet [2] reported the first case of epistatic interaction between genes as they demonstrated that individuals carrying both the Rose-comb and Pea-comb alleles exhibit the walnut-comb phenotype ( Figure 1 ) . Rose-comb has been described in many breeds and shows extensive phenotypic variability ( Figure S1 ) . Most attention has been paid to variation in surface characteristics and texture , angle and number of posterior spikes [3]–[5] . Thus , Rose-comb variability indicates that comb morphogenesis is influenced by several genes and represents an excellent model to study interactions between developmental genes . Numerous reports have documented reduced male fertility associated with the Rose-comb allele [6]–[8] . Gradually it was elucidated that defective sperm motility in roosters homozygous for Rose-comb is the cause of the observed poor fertility and duration of fertility [9]–[15] . This reduced motility is thought to result in sperm from a homozygous Rose-comb rooster ( RR ) being outcompeted by sperm from roosters carrying the wild-type allele ( Rr and rr ) in promiscuous mating or heterospermic insemination experiments [16] , [17] . Heterozygous Rose-comb roosters show normal fertility and transmit their Rose-comb and wild-type alleles to equal number of progeny . Fertility in the hen has been shown to be unaffected by her genotype at the R locus [18] . The unaffected fertility of the heterozygous rooster , combined with the effects of sperm competition , has historically confounded breeders' attempts to establish flocks that breed true for Rose-comb , resulting in an equilibrium of allele frequencies that gives rise to about 15% single-combed chicks in each generation [19] . In the present study , we show that Rose-comb is caused by a large structural rearrangement that leads to transient ectopic expression of an important transcription factor in the chicken , the Mnx-class homeodomain protein MNR2 [20] . This resembles our previous discovery that the Pea-comb mutation constitutes a copy number expansion in intron 1 of SOX5 leading to transient ectopic expression of SOX5 [21] . We also postulate that the sperm motility defect observed in males homozygous for Rose-comb is due to the same structural rearrangement disrupting the CCDC108 gene .
Linkage mapping was performed using a pedigree consisting of two F0 males heterozygous for Rose-comb , sixteen F0 single-combed females , and 383 F1 progeny segregating for Rose-comb ( Rose-comb 50 . 7% , single-comb 49 . 3% ) . Already in 1940 Rose-comb was mapped to chicken linkage group I [22] and in a recent study Rose-comb was assigned to chicken chromosome 7 ( GGA7 ) [23] . Our linkage data , presented in Table 1 , confirmed the assignment to GGA7 and revealed suppression of recombination in Rose-comb heterozygotes as no recombination event was detected over a 7 Mb region , from position 16 . 1 Mb to 23 . 4 Mb on GGA7 , despite the chicken consensus linkage map indicating a distance of 50 cM across this region [24] . This suppression of recombination associated with Rose-comb was confirmed in a second pedigree , a Chinese Silkie x White Plymouth Rock intercross ( Text S1 ) . The observed suppression of recombination suggested that Rose-comb might be associated with an inversion . This hypothesis was strongly supported by a SNP screen using an Illumina 60K SNP array [25] of Rose-comb homozygotes from four Chinese chicken breeds , showing complete homozygosity for all SNPs in the interval 16 , 424 , 096 bp to 23 , 854 , 241 bp despite this region showing normal levels of heterozygosity in wild-type birds ( Figure S2 ) . We searched for an inversion associated with Rose-comb using whole-genome resequencing of a 3 . 9 kb mate-pair library because this approach should precisely predict the location of any inversion breakpoints . The library was prepared from a pool of eight Rose-combed males from the Le Mans breed , all presumed to be homozygous Rose-comb . The library was sequenced to 1× coverage . Bioinformatic analysis of the data revealed aberrant mate pairs consistent with an inversion ( Figure 2 ) . Most aberrant reads ( n = 22 ) indicated a 7 . 38 Mb inversion with breakpoints located approximately at 16 . 50 Mb and 23 . 88 Mb . However , three aberrant mate pairs connected the 16 . 50 Mb region with a region at 23 . 79 Mb , not consistent with a single inversion in all eight Rose-combed individuals included in the pool ( Figure 2 ) . PCR analysis of genomic DNA from Rose-combed individuals confirmed the inversion breakpoints at nucleotide positions 16 , 499 , 781 bp and 23 , 881 , 384–23 , 881 , 392 bp ( Figure 3 ) . A 628 bp gap is predicted around 16 . 50 Mb in the chicken galGal3 assembly . However , sequencing across the gap in the reference sequence bird ( red junglefowl female from the UCD-001 line ) revealed that the gap and an additional 87 bp that together constitute the chr7:16 , 499 , 808–16 , 500 , 522 bp region must be an assembly artefact; the correct sequence of this region has been submitted to GenBank with the accession number JN942757 . The presence of the inversion had an almost perfect association with the Rose-comb phenotype in a PCR-based screen of a large number of individuals from different breeds . No unambiguously single- or Pea-combed individual ( n = 679 ) carried any of the breakpoints , as expected , and almost all Rose- or walnut-combed chickens ( n = 872 ) carried both the 16 . 50 Mb breakpoint and the 23 . 89 Mb breakpoint ( Table 2 ) . However , some individuals from five breeds , that had an unambiguous Rose-comb , ( n = 45 ) showed an aberrant pattern , scoring positive for the 16 . 50 Mb breakpoint but not for the 23 . 89 Mb breakpoint ( Table 2 ) . Therefore we postulated that these birds might carry a second Rose-comb allele , R2 , that evolved from the original Rose-comb allele ( R1 ) by a second rearrangement . Such an event would also be consistent with the presence of the aberrant mate-pair reads connecting the 16 . 50 Mb and 23 . 79 Mb regions ( Figure 2 ) . PCR amplification confirmed the existence of R2 , and the results showed that the allele must have originated by a recombination event between the wild-type allele at position 16 . 50 Mb and the R1 allele at position 23 . 79 Mb in the inverted region ( Multimedia S1 ) . The consequence of this recombination event is that R2 does not carry the entire inversion but instead has two duplicated segments , one 91 kb fragment ( 23 , 790 , 414–23 , 881 , 384 bp ) that represents a remaining fragment of the inverted region together with a small duplicated fragment of 198 bp ( 16 , 499 , 583–16 , 499 , 781 bp ) that is present on both sides of the 91 kb duplication ( Figure 3 ) . Genotyping the eight resequenced Le Mans males revealed an allele composition of 4 R1 , 9 R2 , and 3 r , explaining why a large number of mate pairs over the breakpoint located at 16 . 50 Mb was found , as that is present in both R1 and R2 , while few reads spanning the breakpoint at 23 . 89 Mb were found , as that is only present in R1 . The genome assembly is rich in gaps around 23 . 79 Mb , making it difficult to map reads that span the breakpoint only found in R2 , explaining the low number of mapped reads spanning that breakpoint . FISH analysis using four different BAC clones was employed to confirm the existence of two distinct Rose-comb alleles ( Figure 4; Figure S3 ) . The BAC clones CH261-95H11 and CH261-5G3 span the inversion breakpoints at 16 . 50 Mb and 23 . 88 Mb , respectively . BAC clones TAM32-24B23 and BW27C3 targeted regions within the inversion . A staining of a metaphase spread from an R1r heterozygote confirmed the presence of a large inversion on chromosome 7 . The FISH staining of an R2r heterozygote metaphase spread was consistent with an altered organization , as the BACs CH261-95H11 , TAM32-24B23 and BW27C3 showed indistinguishable staining for both R2 and r chromosomes , with only two aberrant patterns obtained , one for BAC clone CH261-5G3 that confirmed the translocated duplication of a segment from the 23 . 88 MB region to the 16 . 50 MB region , and the other a slight spatial separation of BAC clone CH261-95H11 , consistent with the insertion of the translocated duplication ( Figure 4 ) . Previous studies of the Rose-comb phenotype have established that it involves both an altered comb morphology , showing dominant inheritance , and reduced male fertility , showing recessive inheritance . The presence of two different Rose-comb alleles facilitates the elucidation of the causal relationship between the observed chromosomal rearrangement and these two different aspects of the Rose-comb phenotype . No obvious phenotypic differences in comb morphology were observed amongst birds carrying R1 and R2 , matched for breed and genetic background ( Figure S1 ) . Thus the critical genetic lesion causing the Rose-comb morphology must be located at the 16 . 50 Mb breakpoint including the 91 kb segment transferred from the 23 . 79–23 . 88 Mb region , because this is the only alteration present in both R1 and R2 . The proximal breakpoint at 16 . 50 Mb is located in the 5′UTR of FKBP7 ( FK506 binding protein 7 ) gene , 72 bp upstream of its start codon . Only 9 bp separate the 5′UTRs of FKBP7 and PLEKHA3 ( Pleckstrin homology domain containing , family A-phosphoinositide binding specific-member 3 ) , placing the breakpoint 42 bp from the 5′UTR , and 150 bp from the start codon of PLEKHA3 ( Figure 3B , Multimedia S1 ) . This rearrangement may alter regulation of PLEKHA3 and FKBP7 in both R1 and R2 alleles . The second proximal breakpoint , unique to R2 , where an inverted duplicated 23 . 88-23 . 79 Mb region is joined with a duplicated 198 bp fragment at 16 . 50 Mb results in the duplication of a part of the ABCB6 ( ATP-binding cassette , sub-family B ( MDR/TAP ) , member 6 ) gene . ABCB6 has not been adequately annotated in the chicken genome , but this translocated duplication does not involve the first few exons , judging from annotated Expressed Sequence Tags ( ESTs ) . Additionally , the region in which it is located is riddled with gaps in the genome assembly , causing the 3′ region of the EST range of the gene to appear truncated . The breakpoint seems to be close to the 3′ end of what may be exon 5 or 6 , duplicating the very end of that exon and the remainder of the gene . This exon fragment copy is in its novel genomic context situated only 8 bp from the duplicated first exon of PLEKHA3 . Whilst this could result in a hybrid transcript , the fact that an intact copy of ABCB6 is present on R1 , R2 and r chromosomes , as well as the presence of the duplicated segment being unique to R2 , make it an unlikely candidate for the Rose-comb phenotype . The distal breakpoint at 23 . 88 Mb is located in intron 3 of CCDC108 ( coiled-coil domain containing 108 ) . This breakpoint disrupts CCDC108 and transfers the first three exons to the proximal breakpoint present in both R1 and R2 . However , due to the intact nature of CCDC108 in R2 it was excluded as causative for the altered comb morphology . The neighbouring gene MNR2 ( MNR2 homeodomain protein ) , located only 3 kb from the inside of the distal inversion breakpoint , is also transferred to the near vicinity of the proximal breakpoint in R1 and R2 . The translocation of the transcription factor MNR2 to a novel genomic context was considered the best candidate for causing altered comb morphology in Rose-comb as it belongs to the Mnx-class of homeodomain proteins which act as transcriptional repressors and specifiers of cell identity [26] . Furthermore , hyaluronan ( HA ) , a major component of the extracellular matrix and the comb in chickens , shows strong accumulation around early MNR2-expressing neurons [27] . Expression analysis of comb tissue from early embryos ( single-combed wild-type and R1R1 homozygotes ) by RT-PCR revealed that PLEKHA3 and FKBP7 were expressed in both wild-type and homozygous Rose-comb tissue , whereas CCDC108 and MNR2 were expressed in Rose-combed but not in wild-type embryos ( Figure S4 ) . The ectopic MNR2 expression was restricted to days E6–E12 of embryonic development . This suggested that the Rose-comb phenotype might be due to ectopic expression of the MNR2 homeodomain protein as a copy of MNR2 is translocated close to the 16 . 50 Mb breakpoint in both R1 and R2 . To further explore this possibility , as well as shed light on the interaction between Rose-comb and Pea-comb causing the walnut-comb phenotype , we performed immunohistochemical staining using an anti-chicken MNR2 antibody and an anti-human SOX5 antibody ( previously used in our characterization of Pea-comb [21] ) in single-combed wild-type , Rose-combed , Pea-combed and walnut-combed embryos ( Figure 5 ) . This analysis revealed transient ectopic expression of MNR2 in Rose-combed embryos , consistent with the results of the RT-PCR analysis . Striking MNR2 expression was observed in a layer of mesenchymal cells located in the area where the comb is developing at day E6 . 5 but not at E9 ( Figure 5C and 5D ) . This pattern of transient ectopic expression resembles that previously reported for SOX5 in Pea-combed embryos , where expression is weak at day E5 , strong at E9 ( Figure 5E and 5F ) and not present at day E12 [21] . Walnut-combed embryos showed transient ectopic expression of both MNR2 and SOX5 as expected from their genotype ( Figure 5G–5H ) . The ectopic expression of MNR2 and SOX5 overlapped only partially , with peak expression occurring first for MNR2 . The data revealed ectopic expression of MNR2 and SOX5 in the same cell type as well as MNR2-SOX5 coexpression within individual cells ( Figure 5I ) . Furthermore , MNR2 expression was observed at E9 together with SOX5 ( Figure 5H ) but this was not the case in the absence of SOX5 expression ( Figure 5D ) suggesting a potential positive interaction between the Pea-comb allele of SOX5 and the Rose-comb allele of MNR2 . Interestingly , at day E6 MNR2 and SOX5 also showed ectopic expression in Rose-combed and Pea-combed birds , respectively , in mesenchyme present in the region where the wattles develop ( Figure 5J–5L ) . The ectopic expression of MNR2 , with its transcriptional repression activities , is likely to change the identity of the mesenchyme underlying both the comb and wattles . Seminal work on comb primordium development has shown that the comb shape is directly dependent on instructive signals derived from the underlying mesenchyme/dermal structures [28]–[30] . This suggests that the comb and wattle tissue share a common developmental pathway and that the structural variants underlying Rose-comb and Pea-comb activate the expression of MNR2 and SOX5 , which both modulate or intercept this pathway . However , wattle phenotype is not altered by the Rose-comb mutation as it is by the Pea-comb mutation ( Figure 1 ) . 5′RACE analysis was performed using tissue from early embryonic comb and from adult testis for the three genes located close to the two R1 inversion breakpoints . Results are summarized in Figure S5 . The samples were from single-combed wild-type birds and R1R1 homozygotes . A full length PLEKHA3 transcript ( denoted PLEKHA3a in Figure S5 ) was expressed in both tissues and in both genotypes . A PLEKHA3-CCDC108 hybrid transcript ( PLEKHA3b ) corresponding to exons 1 and 3 from CCDC108 and exons 2–8 from PLEKHA3 was found in Rose-comb testis ( Figure S6 ) . The FKBP7 transcripts showed no difference between genotypes but different splice forms were expressed in comb and testis . Full-length CCDC108 transcripts were only found in wild-type testis ( Figures S5 and S6 ) , whereas two very similar hybrid transcripts lacking the first three exons of CCDC108 were expressed in both R1R1 testis and comb tissue ( Figures S5 and S6 ) . Rose-comb is associated with reduced male fertility in homozygotes due to low sperm motility [12] , [17] , [18] . However , as R1 is the more common Rose-comb allele ( Table 2 ) we wanted to investigate if the fertility effect is also associated with the R2 allele discovered in the present study . A preliminary study to address this question was carried out by mating single-combed and Rose-combed roosters of several genotypes ( R1R1 , R2R2 , R1r and rr ) to single-combed and Rose-combed hens . The data were consistent with reduced male fertility in R1R1 males , as expected , but there were no signs of reduced fertility in R2R2 homozygotes ( Text S2; Table S1 ) . Thus , the deleterious effect on male fertility appears to be restricted to the R1 allele , suggesting that the lesion causing this phenotypic effect is located at the 23 . 88 Mb breakpoint ( Figure 3 ) . The obvious candidate for this effect is the disruption of the CCDC108 transcript that leads to the expression of a truncated transcript in testis ( Figures S5 and S6 ) .
The present study illustrates several striking features of the genetic diversity present in domestic animals . Firstly , the R2 allele exemplifies the evolution of alleles by two or more consecutive mutations . Other examples include the Dominant white allele in pigs which involves a 450 kb duplication encompassing the entire KIT gene combined with a splice mutation in one of the duplicated copies [31] and black spotting in pigs which is determined by the combined effects of two mutations in MC1R , a missense mutation associated with black colour and a somatically unstable two base-pair insertion [32] . Secondly , it represents a new example of how structural rearrangements have contributed to rapid phenotypic evolution observed in domestic animals [33] , [34] . The majority of the structural changes reported to be associated with phenotypic effects , like the effects of R1 and R2 on comb morphology , constitute cis-acting regulatory mutations . The altered configurations of regulatory elements on the rearranged chromosome lead to altered gene expression patterns . It appears plausible that structural rearrangements similar to those that affect comb development in chickens have also contributed to phenotypic evolution in natural populations , including human populations . There are several mechanisms by which genomic rearrangements are thought to occur , involving either double strand break repair via primarily non-homologous mechanisms or homology mediated replication and recombination based processes [35] , [36] . At the R1 and R2 16 . 50 Mb breakpoint there is only a single base pair sequence overlap , while at the R1 23 . 88 Mb breakpoint there is a 7 bp overlap with one mismatch . Although it is not possible to determine the exact mechanism by which the original inversion occurred , it is likely that microhomology at the 23 . 88 Mb breakpoint was involved in the generation of the R1 allele . The generation of the R2 allele occurred by recombination between a wild-type chromosome and an R1 chromosome ( Multimedia S1 ) . On the wild-type chromosome this event occurred 198 bp upstream of the 16 . 50 Mb breakpoint , on the R1 chromosome the recombination event occurred 91 kb into the inversion from the 16 . 50 Mb breakpoint . This resulted in the duplication of the 91 kb portion of the R1 chromosome , including the R1 proximal breakpoint and the 199 bp fragment of the wild-type chromosome , effectively inserting the duplicated sequence at 16 . 50 Mb into a wild-type chromosome . Analysis of the R2 recombination event breakpoint shows 2 bp of sequence homology , again suggesting that sequence microhomology was likely involved . That both events introduced breakpoints around 16 . 50 Mb , spaced only 198 bp apart , suggests that some characteristic of this region may predispose it to such events . More than half of the loci identified in human genome-wide association analyses do not overlap coding regions [37] , implying that they reflect regulatory polymorphisms . The present study illustrates how challenging it can be to reveal the molecular mechanism underlying even a simple monogenic trait in a model organism such as the chicken . We were able to demonstrate transient ectopic expression of MNR2 during a narrow period of embryonic development because immunohistochemistry provided the spatial resolution allowing the detection of ectopic expression in a subset of cells within the affected tissue ( Figure 5 ) . We postulate that the well-established association between the Rose-comb phenotype and reduced sperm motility is restricted to the R1 allele and that it is caused by the disruption of the CCDC108 transcript . The predicted CCDC108 protein sequence in chickens shows 49% amino acid identity with human CCDC108 ( HomoloGene:28093; www . ncbi . nlm . nih . gov ) . CCDC108 has an unknown function , according to the UNIPROT database ( www . uniprot . org/uniprot/Q6ZU64 ) , it is a single pass membrane protein composed of 1925 amino acids , containing one MSP ( major sperm protein ) domain . The MSP domain is present in major sperm proteins and in sperm specific proteins ( SSPs ) found in various nematodes as well as in the mammalian Motile sperm domain-containing proteins-1 , -2 and -3 ( MOSPD1-3 ) . All MSP , SSP and MOSPD proteins are small , with a size range of 107–518 amino acids , and thus much smaller than CCDC108 . Mouse Ccdc108 is expressed in testis and shows differential expression during progression of spermatogenesis [38]; expression is absent in juvenile mice but is turned on when male mice reach sexual maturity ( www . ncbi . nlm . nih . gov; GEO profiles GDS606/164004_at/Ccdc108/Mus musculus ) . Furthermore , by using reciprocal best-hit protein BLAST searches , putative CCDC108 orthologs can be found in many organisms , including in deep branches on the tree of life . One such orthologous protein has been annotated in Chlamydomonas algae as an axonemal protein named Flagellar Associated Protein 65 ( FAP65 ) [39] . FAP65 expression is strongly induced after deflagellation , when cells regenerate their flagella . This sequence homology suggests that CCDC108 is part of the sperm flagellum and thus that a disruption of CCDC108 function may lead to sperm motility defects as observed in Rose-comb R1R1 homozygotes . This study establishes CCDC108 as a candidate gene for sperm motility disorders in humans . The present study strongly suggests that Rose-comb and Pea-comb are caused by transient ectopic expression of two potent transcription factors , MNR2 and SOX5 , respectively . However , at present we cannot exclude the possibility that altered expression levels of FKBP7 or PLEKHA3 , located on either side of the proximal breakpoint , may have some impact on the comb phenotype . However the ectopic MNR2 expression in the developing embryo is by far the most striking molecular consequence of the Rose-comb inversion . The exact downstream mechanisms leading to the altered comb morphology remain undetermined . Comb tissue is composed of layers of epidermis , dermis and central connective tissue comprising primarily collagen and hyaluronan ( HA ) [40] . The previous report that there is a strong accumulation of HA around early MNR2-expressing neurons [27] may be relevant for the Rose-comb phenotype , but perhaps more importantly , MNR2 acts as a repressor and specifier of cell identity [26] . SOX5 also has an established functional role that makes sense in relation to the altered comb morphology observed in Pea-combed and walnut-combed birds . SOX5 contributes to chondrogenesis , together with SOX6 and SOX9 it activates specific genes during embryonic cartilage formation [41] , and has a repressive role in oligodendrogenesis during neural development [42] . A fascinating observation is that both the Rose-comb mutations and the Pea-comb mutation give rise to ectopic expression in the area of the developing comb , leading to altered comb morphology in both mutants , as well as in the wattle area , which only leads to altered morphology in birds carrying Pea-comb ( Figure 1 ) . The fact that the transient ectopic expression of MNR2 and SOX5 apparently occurs in the same population of mesenchymal cells , and with at least partially overlapping expression in individual cells in the comb primordium , provides a reasonable explanation of why the combined effect of the two mutations leads to the formation of the walnut-comb . Thus , 104 years after Bateson and Punnett [2] reported the first example of epistatic interaction between genes , we can now provide a molecular explanation for their seminal observation .
All animal work has been conducted according to relevant national and international guidelines . Linkage mapping was carried out using a pedigree consisting of two heterozygous R1r male parentals , each mated with eight homozygous rr females , resulting in 383 progeny segregating for Rose-comb . The Rose-combed roosters were from an INRA ( French National Institute for Agricultural Research ) resource population , with the R1 allele having been derived from the French breed Charollaise . The wild-type single-combed hens were from another INRA resource population line . DNA samples from various chicken breeds were genotyped for the R1 , R2 and wild-type alleles . These included samples collected as part of the AvianDiv project [43] , from resource populations at INRA , and 31 different breeds of Chinese chickens collected by Institute of Poultry Science , Chinese Academy of Agricultural Sciences . Blood samples from privately owned Icelandic chickens were obtained at eight different locations in the South-West of Iceland by a veterinarian with permission from owners . Genomic DNA from the reference red junglefowl bird was kindly provided by Dr . J . B . Dodgson . Linkage analysis was performed by genotyping two microsatellites and nine SNPs from chromosome 7 using standard procedures . Custom TaqMan SNP Genotyping Assays ( Applied Biosystems ) were designed by ABI , other primers were designed with the Primer3Plus webtool ( http://www . bioinformatics . nl/cgi-bin/primer3plus/primer3plus . cgi ) . See Table S2 for primer and probe sequence information . Linkage analysis was performed with the Crimap software ( version 2 . 4 ) [44] . DNA from eight Rose-combed males from the Le Mans breed , all presumed to be homozygous for Rose-comb , were pooled . Whole genome resequencing data from a pool of Rose-combed Silkie chickens , and another pool of single-combed White Leghorns were obtained at later time points and included in this study to verify the results obtained from the Le Mans pool . A sequencing library was generated for the Le Mans sample using a Mate-pair SOLiD3 protocol and sequenced on SOLiD v . 3 ( Life Technologies , Carlsbad , USA ) . The White Leghorn library was generated using a Mate-pair SOLiD3+ protocol and sequenced on SOLiD 3+ . The Silkie Library was generated using a Mate-pair SOLiD5500 protocol and sequenced on SOLiD5500XL . The Le Mans , White Leghorn and Silkie reads ( 2×50 bp mate-pair reads ) were mapped to the chicken genome ( WUGSC 2 . 1/galGal3 ) reference assembly using the software CoronaLite v0 . 4r2 , Bioscope v1 . 0 . 1 and LifeScope v2 . 0 , respectively , with average insert sizes estimated as approximately 3 . 9 , 3 . 1 and 2 . 6 kb and average read depth approximately 1× , 10× and 20× over the chicken genome . Mapping distances between mate-pairs were used to detect structural variations in relation to the reference assembly . All library kits , alignment software and massively parallel sequencing equipment were used according to the manufacturer's instructions ( Life Technologies , Carlsbad , USA ) . Heterozygous Rose-combed embryos ( R1r and R2r ) were produced from parental stock maintained at INRA . The R1 allele originated from Belgian Barbu d'Anvers and the R2 allele from French Alsacienne . BAC clones were chosen considering their position in the chicken genome sequence ( Table S3 ) . BW27C3 comes from the Wageningen library [45] . TAM32-24B23 was ordered from TAMU ( Texas A&M BAC Libraries , USA ) . CH261-5G3 and CH261-95H11 were ordered from the Children's Hospital Oakland Research Institute in Oakland ( CHORI ) , California , USA . BAC clones were grown in LB medium with 12 . 5 µg/ml chloramphenicol according the instructions of the providers . The DNA was extracted using the Qiagen plasmid midi kit . FISH was carried out on metaphase spreads obtained from fibroblast cultures derived from 7 days old embryos , arrested with 0 . 05 µg/ml colcemid ( Sigma ) and fixed by standard procedures . The FISH protocol is derived from Yerle et al . [46] . Two-colour FISH was performed by labelling 100 ng of each BAC clone with alexa fluorochromes ( ChromaTide Alexa Fluor 488-5-dUTP , Molecular probes; ChromaTide Alexa Fluor 568-5-dUTP , Molecular Probes ) by random priming using the Bioprim Kit ( Invitrogen ) . The probes were purified using spin column G50 Illustra ( Amersham Biosciences ) . Probes were ethanol precipitated together and hybridised to the metaphase slides for 17 h at 37°C in the Hybridizer ( Dako ) after denaturation for 8 min at 72°C . Chromosomes were counterstained with DAPI in antifade solution ( Vectashield with DAPI , Vector ) . The hybridised metaphases were screened with a Zeiss fluorescence microscope . A minimum of twenty spreads was analysed for each experiment . Spot-bearing metaphases were captured and analysed with a cooled CCD camera using Cytovision software ( Applied Imaging , Leica-Microsystem ) . Images were formatted , resized and arranged for publication using Adobe Photoshop and Adobe Illustrator . A set of five PCR primers that together will amplify a series of specific bands over each of the five breakpoints was designed for genotyping the R1 , R2 and r alleles . Primer and protocol information are in Table S4 . Gel image for the six possible genotypes is presented in Figure S7 . Comb tissue was collected from homozygous ( R1R1 ) and heterozygous ( R1r ) Rose-combed birds as well as homozygous ( rr ) single-combed wild-type birds . Comb tissue was sampled at embryonic ( E ) days 6 , 7 , 8 , 9 , 10 , 11 , 12 , 15 and 19 . Testis tissue was sampled from adult roosters at day 200 . Samples from three birds of each type were collected and stored in RNAlater ( Ambion ) . RNA was extracted using RNeasy Mini kit ( Qiagen ) . cDNA was synthesized with 1 µg of RNA using oligo ( T ) primer . Primers spanning introns were used in RT-PCR . The 5′ and 3′ RACE were performed using GeneRacer Kit ( Invitrogen ) . Homozygous Rose-combed Alsacienne , single-combed INRA resource population , homozygous Pea-combed Cheptel and heterozygous walnut-combed Alsacienne x Cheptel embryos were used . Heads from staged embryos were fixed in 4% paraformaldehyde in phosphate buffered saline ( PBS ) for one hour at 4°C . Fixed heads were incubated overnight in 30% sucrose in PBS at 4°C , embedded in OCT freezing medium ( Tissue-Tek , Sakura ) , frozen and sectioned in a cryostat . Cross sections , 10 µm thick , were collected on glass slides ( Super Frost Plus , Menzel-Gläser ) . The sections were rehydrated in PBS for 5 min and then blocked for one hour in PBS containing 1% fetal calf serum , 0 . 1% Triton-X and 0 . 02% Thimerosal . The antibodies MNR2 ( Developmental studies hybridoma bank , 81 . 5C10 ) and SOX5 ( Abcam , a_6226041 ) were diluted 1∶250 and 1∶1000 respectively in blocking solution and incubated on the slides overnight at 4°C . The secondary antibodies ( Invitrogen ) were incubated at room temperature for two hours at a 1∶1000 dilution in blocking solution . Samples were analysed using a Zeiss Axioplan2 microscope equipped with Axiovision software . Images were formatted , resized , enhanced and arranged for publication using Axiovision and Adobe Photoshop . Information on the chicken genome sequence is available at http://www . genome . ucsc . edu . The sequence data presented in this paper have been submitted to GenBank with accession numbers JN942757–JN942760 , JN880446 , JN880447 , JQ004983 , and JQ004984 . | Comb morphology is a trait that shows considerable variability among domestic chickens . The Rose-comb mutation causes a drastically altered shape of the comb , whereas the Pea-comb mutation leads to a considerable reduction in the size of the comb . The combined effect of Rose-comb and Pea-comb is a comb shaped like a walnut , and the phenotype is consequently named walnut-comb . Both Pea-comb and Rose-comb are caused by structural changes in the genome leading to altered expression of important transcription factors . In a previous study we showed that Pea-comb is caused by misexpression of SOX5 during the development of the comb . In this study we report that Rose-comb is caused by a large inversion on chicken chromosome 7 . The inversion moves the MNR2 gene to a new genomic location . This leads to misexpression of MNR2 during comb development , similar to the defect causing Pea-comb . Roosters that are homozygous for the Rose-comb inversion show poor sperm motility , and our results suggest that this is caused by the disruption of the CCDC108 gene that is located at one of the inversion breakpoints . CCDC108 is well conserved between chickens and humans , and this study establishes CCDC108 as a candidate gene for sperm motility disorders in humans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"animal",
"genetics",
"functional",
"genomics",
"gene",
"function",
"molecular",
"genetics",
"structural",
"genomics",
"veterinary",
"science",
"veterinary",
"medicine",
"chromosome",
"biology",
"gene",
"expression",
"comparative",
"genomics",
"biology",
"genetics",
"geno... | 2012 | The Rose-comb Mutation in Chickens Constitutes a Structural Rearrangement Causing Both Altered Comb Morphology and Defective Sperm Motility |
The molecular basis for the formation of functional , higher-ordered macro-molecular domains is not completely known . The Kaposi’s Sarcoma-Associated Herpesvirus ( KSHV ) genome forms a super-molecular domain structure during latent infection that is strictly dependent on the DNA binding of the viral nuclear antigen LANA to the viral terminal repeats ( TR ) . LANA is known to form oligomeric structures that have been implicated in viral episome maintenance . In this study , we show that the LANA oligomerization interface is required for the formation of higher-order nuclear bodies that partially colocalize with DAXX , EZH2 , H3K27me3 , and ORC2 but not with PML . These nuclear bodies assemble at the periphery of condensed cellular chromosomes during mitotic cell division . We demonstrate that the LANA oligomerization interface contributes to the cooperative DNA binding at the viral TR and the recruitment of ORC to the viral episome . Oligomerization mutants failed to auto-regulate LANA/ORF73 transcription , and this correlated with the loss of a chromosome conformational DNA-loop between the TR and LANA promoter . Viral genomes with LANA oligomerization mutants were subject to genome rearrangements including the loss of subgenomic DNA . Our data suggests that LANA oligomerization drives stable binding to the TR and formation of an epigenetically stable chromatin architecture resulting in higher-order LANA nuclear bodies important for viral genome integrity and long-term episome persistence .
Kaposi’s Sarcoma Associated Herpesvirus ( KSHV ) is a human gammaherpesvirus responsible for Kaposi’s Sarcoma ( KS ) , Pleural Effusion Lymphoma ( PEL ) , and multicentric Castleman’s Disease ( mCD ) ( reviewed in [1 , 2] ) . KSHV is a large double-stranded DNA virus that establishes life-long latent infection in B-lymphocytes and potentially other cell types , such as mesenchymal stem cells and endothelial cells . Most KSHV-associated cancers harbor latent viral genomes in the nucleus of tumor cells . The latent genomes form covalently closed circular genomes , termed episomes , that establish a local chromatin structure similar to that of cellular genomic DNA ( reviewed in [3–5] ) . During latency , viral gene expression is restricted to a small set of viral genes . Viral DNA replication occurs coordinately with host cell division cycle utilizing host cellular DNA replication machinery . Stable maintenance of the latent episome requires the regulated assembly of chromatin factors on the viral genome to enable selective expression of latent , but not lytic viral genes . Furthermore , the viral episome must replicate on average once per cell cycle and be transmitted faithfully to each daughter cell to maintain stable copy number in each daughter cell . The KSHV encoded Latency Associated Nuclear Antigen ( LANA ) is a multifunctional DNA binding protein that is essential for stable maintenance of KSHV episomes during latent infection [6–8] . The LANA DNA binding domain shares structural homology to the DNA binding domains of EBV EBNA1 and HPV E2 proteins ( [9–12] reviewed in [13] ) . LANA binds directly to three recognition sites in the terminal repeats ( TR ) of the KSHV genome [9] . These sites represent a minimal origin of plasmid DNA replication [14] , and at least two or more TRs are required for episome maintenance [15] . LANA is required for the efficient recruitment of cellular replication initiation factors , such as the Origin Recognition Complex ( ORC ) and the Mini Chromosome Maintenance ( MCM ) complex to the viral TR region [16 , 17] . LANA binding to the TR is thought to influence the chromatin organization of the viral episome through recruitment of ORC and other chromatin-regulatory proteins [16 , 18 , 19] . LANA can also interact with core histones H2A/H2B through its amino terminal domain that facilitates attachment of the virus genome to a metaphase chromosome and is essential for viral genome transmission during cell division [20–23] . LANA is also thought to influence the higher-order chromatin and chromosomal structure of the KSHV episome during latency . Fluorescent light microscopy reveals that LANA forms large aggregate structures , often termed LANA speckles , that colocalize with KSHV episomes in the nucleoplasm during interphase and with telomeric and centromeric regions of the metaphase chromosome in mitosis [11 , 15 , 24 , 25] . Some of these LANA structures were found to colocalize with heterochromatin-associated proteins , such as DAXX [26 , 27] and the Polycomb-associated histone H3K27 methyltransferase EZH2 [28] . Super resolution imaging studies indicate that LANA coordinates a higher-ordered architecture at the TR of stable episomes [29] . Live cell imaging studies indicate that KSHV episomes can form large , multigenome clusters connected through LANA bivalent C-terminal DNA binding and N-terminal histone tethering activities [30 , 31] . The LANA DNA binding domain ( DBD ) by itself can form a variety of higher-order oligomeric structures , as determined by X-ray crystallography and biochemical methods [10 , 11] . Mutations in LANA that disrupt these higher order oligomers in vitro inhibited cooperative DNA binding and plasmid maintenance , but had no observable effect on DNA binding to single LANA binding sites [10] . It remains unclear whether molecular oligomerization of LANA DBD is directly associated with the higher order nuclear structures observed by microscopy methods [15 , 24 , 26 , 27 , 29 , 30] . Here , we examine the role of the LANA DBD oligomerization interface in the assembly of these higher-order structures observed by fluorescence light-microscopy . In addition , we assessed if this process influences viral gene expression , chromosome conformation , and viral genome stability . Our data suggest that LANA DBD oligomerization provides the molecular basis for KSHV chromosome conformation and super-molecular nuclear bodies required for stable episome maintenance during latency .
LANA has been shown to form multiple aggregate structures of various sizes in the nucleus of latently infected cells [11 , 15 , 24 , 25] . These structures have also been shown to colocalize with several other nuclear factors , including DAXX [26] and EZH2 [28] . We confirmed these findings in BCBL1 pleural effusion lymphoma cell line ( Fig 1A and 1B ) . To further investigate the formation and transmission of these LANA structures in living cells , we fused an amino-terminal RFP to the N-terminus of LANA in the infectious KSHV bacterial artificial chromosome ( BAC ) clone BAC16 ( Fig 1C ) . We generated stable iSLK cell lines carrying latent KSHV BAC16 genomes that constitutively express RFP-LANA . RFP-LANA in iSLK cells formed similar nuclear body patterns as was observed for endogenous LANA in BCBL1 cells , including the colocalization with DAXX ( Fig 1D ) and EZH2 ( Fig 1E ) . We also found that RFP-LANA partially colocalizes with H3K27me3 ( Fig 1F ) , consistent with the role of EZH2 in histone methylation . While DAXX commonly colocalizes with the anti-viral protein PML and PML-nuclear bodies ( PML-NBs ) , we found that RFP-LANA did not colocalize with PML ( Fig 1G ) . These findings indicate that RFP-LANA expressed from KSHV BAC16 in iSLK cells forms similar nuclear bodies to that observed in naturally occurring latently infected PEL cells , and that these LANA bodies are distinct from PML-NBs . To investigate the stability and transmission of LANA nuclear bodies through the cell cycle , we used live cell confocal microscopy of RFP-LANA in stable iSLK cells carrying KSHV BAC16 expressing RFP-LANA ( Fig 2 ) . During cell division , RFP-LANA bodies assemble on the metaphase plate and segregate with equal partitioning to daughter cells ( Fig 2A ) . Within 20 min after metaphase LANA clusters reorganized in a distribution pattern strikingly similar to the parent cell , typically forming a ring of partially connected bodies along the nuclear periphery ( Fig 2A ) . High resolution 3D reconstruction of the Z-stacked confocal images suggest that these nuclear bodies are heterogeneous in size ( Fig 2B ) . Visualization of the division cycle from three different viewpoints revealed that the ring of LANA nuclear bodies undergoes an orthogonal re-orientation during telophase and the formation of two new rings in similar distribution and planar orientation as the parent cell ( Fig 2C , supplemental movies M1-2 ) . Costaining of mitotic figures with α-tubulin and Dapi further demonstrated that LANA bodies partition coordinately with host chromosomes with typical colocalization at the periphery of condensed metaphase chromosomes ( Fig 2D and 2E ) . These findings indicate that LANA forms highly organized nuclear structures that undergo a coordinated transmission process and stable morphology in recipient daughter cells . Previous structural and molecular genetic studies have demonstrated that amino acids F1037 and F1041 form the oligomerization interface of the LANA DBD ( Fig 3A and 3B ) [10] . Mutation of this oligomerization interface ( F1037A/F1041A ) had no effect on LANA DNA binding to a single LANA binding site ( LBS ) , but were found to disrupt oligomerization and cooperative DNA binding in vitro , and inhibit LANA-dependent plasmid maintenance in vivo [10] . Now , we demonstrate that LANA oligomerization is necessary for functional binding to the TR in vivo ( Fig 3C and 3D ) . To this end , LANA DBD wild type ( WT ) or LANA DBD containing oligomerization interface mutations ( F1037A/F1041A ) ( MT ) were fused to the VP16 transcriptional activation domain and FLAG-tagged . We then assayed these for their ability to bind and activate transcription from a luciferase reporter plasmid with either a single LANA binding site ( 1xLBS ) or three naturally occurring LBS ( 3xLBS ) as found in the TR origin of DNA replication ( Fig 3C ) . We found that LANA DBD WT could activate transcription ~2 fold from 1xLBS and ~20 fold from 3xLBS reporter , while LANA DBD MT failed to activate 1xLBS and activated 3xLBS ~4 fold . This difference was more striking considering that LANA MT was expressed at higher levels than LANA WT , as detected by Western blot ( Fig 3D ) . These findings indicate that LANA oligomerization interface contributes to the functional binding at the LBS in living cells and support the model that cooperative DNA binding is important for LANA function at the TR . To investigate whether LANA oligomerization is required for the higher-order nuclear bodies observed by light microscopy , we engineered the F1037A and F1041A mutations into RFP-LANA BAC16 construct ( Fig 4A ) . We generated at least two independent iSLK cell lines for each bacmid . We observed a significant difference in the frequency and abundance of nuclear bodies . LANA MT formed more diffuse nuclear fluorescence patterns compared to LANA WT ( Figs 4B and S1A ) . LANA nuclear bodies were observed during primary infection of SLK ( S1B Fig ) and HUVEC ( S1C Fig ) cells . LANA WT formed visible foci in the nuclei of infected cells within 48 hrs post-infection . In contrast , LANA MT formed more diffuse and less intense fluorescence signals at the same time point ( S1B and S1C Fig ) . In stable iSLK cells , RFP-LANA WT colocalized with DAXX and EZH2 in nuclear foci , while RFP-LANA MT appeared as diffuse pattern with dispersed colocalization ( Fig 4C and 4D ) . LANA oligomerization mutation did not disrupt the ability of LANA to co-immunoprecipitate with DAXX ( S2 Fig ) , consistent with other studies mapping a DAXX interaction interface to regions outside the LANA DBD [27] . These finding indicate that mutations in the LANA oligomerization interface ( F1037A/F1041A ) reduce the ability of LANA to form nuclear bodies that colocalize with DAXX and EZH2 in living cells ( S3 Fig ) . To evaluate the potential effects of LANA oligomerization mutants on the KSHV epigenome , we assayed the interaction of several DNA-binding and chromatin regulatory factors with the KSHV episome by chromatin-immunoprecipitation ( ChIP ) assay ( Fig 5 ) . We first investigated LANA binding and observed that oligomerization mutants were significantly impaired for their ability to bind to the KSHV TR ( primer h ) relative to LANA WT ( Fig 5A and 5B ) . This is consistent with the previous finding that mutant LANA is compromised for activating a TR-dependent reporter gene in vivo ( Fig 3 ) . ORC2 , a factor known to be recruited to TR by LANA and functionally important for viral DNA replication and episome maintenance [16] , was completely eliminated from TR in LANA MT relative to WT KSHV genomes ( Fig 5B ) . Consistent with ChIP , we found that ORC2 colocalized with LANA WT foci , but only formed diffuse patterns with LANA MT ( Figs 5C and S3 ) . Histone H3K27me3 was found to be enriched at the lytic control region ( regions a , c , d ) , but enrichment at this region was not significantly affected by LANA oligomerization mutants . On the other hand , intermediate levels of H3K27me3 found at the latency control region ( primers e ) were significantly reduced by LANA MT ( Fig 5B ) . The pattern of binding of CTCF , H3K4me3 , and total H3 were similar to that reported for BCBL1 [32] , but this binding was not significantly affected by LANA oligomerization mutants ( S4 Fig ) . RAD21 trended to be generally reduced throughout the KSHV genome in LANA MT relative to WT , although the levels of binding were relatively low throughout the genomes in iSLK cells ( S4 Fig ) . These findings indicate that LANA oligomerization mutants have defects in binding to the TR , recruiting ORC2 to the TR , and maintaining H3K27me3 at the latency control region . LANA oligomerization domain mutants were next assayed for LANA protein and RNA expression ( Fig 6 ) . iSLK cells carrying oligomerization domain mutants showed an increase in the total abundance of LANA protein , relative to cells carrying wildtype LANA for two independent lines of each WT or MT LANA ( Fig 6A ) . On the other hand , LANA MT containing cells produced lower levels of viral lytic proteins ORF45 and ORF50 after doxycycline induction of RTA in iSLK cells ( S5A Fig ) . These same cell lines were assayed by RT-qPCR for the expression of several viral genes including the latency transcripts for ORF71 , ORF72 , and ORF73 ( Fig 6B ) , and lytic transcripts for ORF50 , ORF45 , and PAN ( S5B Fig ) . Consistent with Western blot results , MT1 and MT2 produced higher levels of ORF73 , but not ORF71 or ORF72 relative to WT controls ( Fig 6B ) , and reduced levels of ORF45 transcript ( S5B Fig ) . These findings suggest that LANA oligomerization mutants fail to negatively auto-regulate ORF73/LANA transcription . To investigate potential mechanisms for LANA autoregulation , we used chromatin conformation capture ( 3C ) to determine if the LANA binding at the TR may contact potential regulatory regions controlling the LANA transcript . We have previously observed a strong 3C DNA loop between the latency control and lytic control regions that was dependent on CTCF and cohesin subunit RAD21 in BCBL1 cells [33] . We found that this 3C interaction was weak in iSLK cells , but completely undetectable in genomes with LANA oligomerization mutants ( S6 Fig ) . To determine if there were 3C interaction with the TR element and the LANA promoter , we positioned a 3C anchor primer adjacent to the BamHI site closest to the KSHV TR ( Fig 6C ) . We then assayed whether this anchor primer could form any detectable loops with other regions of the KSHV genome . We found that the TR anchor formed a robust interaction with the region at 130471 , which is situated ~3 kb upstream from the ORF73 transcription start site ( Fig 6C ) . Importantly , this TR-mediated interaction was significantly reduced by LANA oligomerization mutants MT1 or MT2 relative to LANA WT . These findings suggest that LANA oligomerization is important for a viral chromosome conformation that involves contacts between the TR and other regulatory regions . Since LANA oligomerization mutants have defects in long-term episome maintenance , we considered the possibility that mutant viral genomes may be genetically unstable . We therefore quantified by qPCR a series of viral genomic regions in total DNA isolated from iSLK clones carrying MT and WT LANA ( Fig 7 ) . We identified a large region spanning from base pair 1 to 110 , 000 in the virus genome that had a significantly reduced copy number relative to bacmid hygromycin gene sequence ( Fig 7A and 7C ) . This region contains lytic control genes and duplicated lytic origins . We did not detect any consistent loss of copy number at the latency control region and TR proximal regions ranging from 126 . 5–137 . 3 kb of the viral genome . Importantly , we did not detect any copy number variation in the starting bacmid DNA for MT1 or MT2 relative to the WT bacmid used for generating iSLK stable cell lines ( Fig 7B ) . Sequencing of these starting bacmids did not reveal any deletions or mutations other than the expected point mutations in the LANA DBD ( S1 Data ) . PFGE analysis of these cells indicated that episomal genomes were maintained in all cell lines , but gross rearrangements relative to parental and WT genomes were observed with MT1 and MT2 ( Fig 7D and 7E ) . Similar genetic instability and viral phenotypes were observed in an independently derived pair of cell lines using a different pair of bacmid clones ( WTgfp and MTgfp ) , indicating that these phenotypes are attributable to LANA oligomerization mutation followed by propagation in cell culture ( S7–S9 Figs ) . These findings suggest that LANA oligomerization mutants fail to maintain the integrity of the complete KSHV genome , with loss of DNA encompassing the lytic origins and lytic control regions after prolonged selection in cell culture .
Formation of complex , higher-ordered structures from simple repetitive units is a common theme in biological systems [34] . LANA is a multifunctional protein that binds cooperatively to three tandem recognition sites in a single viral TR . The viral TR is an ~850 bp element that is tandemly repeated as many as 35 times in some viral genomes . Multiple tandem copies of TR are required for LANA-dependent long-term stable episome maintenance of the complete viral genome [15 , 35] . How LANA builds higher-ordered functional structures from this repetitive binding block is not completely clear . Here , we provide evidence that the simple homo-oligomeric interactions of the LANA DBD drive formation of higher-order chromatin architecture and nuclear domain structures required for stable episome maintenance of KSHV genomes during persistent latent infection ( Fig 8 ) . LANA can self-assemble as a variety of different oligomeric forms in X-ray crystallographic lattices , including decamers , pentamers , and spirals [9–12] . LANA can also form continuous filamentous oligomers when bound to non-specific DNA in electron microscopy studies [9] . There is evidence that LANA can initiate additional DNA interactions through a basic patch on the back surface of the sequence-specific DNA binding domain [9] . LANA is also known to have an N-terminal nucleosome interaction domain important for tethering to metaphase chromosomes and essential for episome maintenance [20] . These multiple DNA and chromatin interactions by LANA contribute to its ability to form complex architectural structures . Previous studies have shown that mutations in the LANA oligomerization interface have no measurable effect on LANA-DNA binding to a single LBS , but did have defects in cooperative binding to larger DNA containing multiple LBS as found in native TR elements [10] . We now find that LANA oligomerization mutants are compromised for binding to KSHV TR in living cells . This suggests that oligomerization is required for stable binding to multiple LBS in the TR . LANA oligomerization and cooperative binding may be necessary to overcome the energetic barriers of DNA and nucleosome structures that prohibit monomeric LANA binding . Our data suggest that cooperative assembly and precise geometry of LANA oligomers at TR is critical for the formation of these higher order structures and their associated functions . The KSHV genome can form complex DNA interactions , including a DNA loop between the latent and lytic control regions that are mediated by CTCF-cohesin interactions [33 , 36] . We now report the identification of an additional 3C DNA loop between the TR and the region upstream of the LANA promoter . Supporting evidence for this interaction is provided by ChIP-Seq experiments showing a high affinity interaction with the TR and a secondary , lower affinity LANA interaction with the region upstream of LANA transcription start site [37 , 38] . The TR anchored DNA loop was disrupted by mutations in the LANA oligomerization interface , suggesting that higher-order LANA oligomerization is required for loop formation . Since LANA oligomerization was also required for efficient TR binding as well as ORC2 and RAD21 recruitment , it is likely that these factors and others also contribute to the formation of a stable DNA conformation . The function of this LANA-dependent loop may be in the auto-regulation of the LANA transcript . We observed that the LANA/ORF73 transcript and protein were up-regulated in the presence of mutations that abrogate LANA oligomerization . We suspect that LANA DNA loop formation between TR and LANA transcriptional regulatory region plays a role in the auto-repression of LANA transcription . Similar autoregulation is observed for EBNA1 binding to its transcription initiation site at the Qp promoter in the EBV genome [39 , 40] . Transcriptional autoregulation has been implicated in the copy number control mechanisms for several plasmid systems , including the yeast 2 micron plasmid [41] , and may play an important role in regulating KSHV episome copy number and latency control . Genomes containing mutant LANA failed to maintain genetic stability over time , with a loss of viral DNA from regions more distal to the TR and latency control region . One potential explanation for this loss of genomic DNA is that TR fails to recruit ORC and function efficiently as a replication origin . Inefficient origin firing could result in the incomplete replication of the viral genome over multiple cell divisions . We also observed that the boundary for genetic loss appears to coincide with the TR and the right lytic origin ( KSHV genome coordinates ~1–110 , 000 ) , suggesting that lytic origin activation may have destabilized this region of the genome . Although the mechanism for loss of genetic material is not completely understood , these findings underscore the crucial role of LANA oligomerization in maintaining viral genome integrity . The super-molecular structures formed by LANA and KSHV episomes can be readily visualized by light microscopy as LANA nuclear bodies . Large LANA bodies were observed in a large percentage of cells carrying LANA WT , but to a much lesser extent in oligomerization defective LANA MT . This suggests that LANA oligomerization is the driving force for the formation of these higher order structures . These large structures form at much higher frequency in cells where KSHV episomes are stably maintained , suggesting that additional epigenetic or stochastic factors contribute to the formation of a stable LANA nuclear body . We observed a high colocalization of LANA nuclear bodies with DAXX . These bodies lacked PML , suggesting they are distinct from traditional PML-NBs . DAXX has been implicated in histone H3 . 3 chaperone activity and transcription repression [42–44] , but may also play a role in macromolecular protein complex assembly [45] . We suggest that the function of DAXX in these LANA nuclear bodies is to facilitate formation and stabilize these large multi-protein , nucleic acid complexes . We also observed partial colocalization with EZH2 and H3K27me3 , consistent with other studies showing the importance of polycomb repression in maintaining KSHV latency [28 , 46–48] . We also observed that ORC2 colocalized with a large fraction of LANA bodies . Numerous other factors have been shown to colocalize with LANA , including γH2AX [49] , H3K9me3[50] , KDM3A[51] , RFC[52] , NRF2 [53] , as well as DAXX [26 , 27 , 54] . Whether these other factors contribute to the formation of LANA nuclear bodies is not known . LANA nuclear bodies are typically only observed when LANA is associated with viral genomes [15] . Super-resolution microscopy of LANA interaction with 2xTR DNA revealed LANA can nucleate a hub at the TR with multiple interactions to host chromatin and bending of nucleosomal DNA consistent with activated chromatin [29] . Live cell imaging using fluorescently tagged episomes indicate that KSHV genomes form multicopy clusters , that tend to segregate to a single daughter cells [30] . Our live cell imaging of RFP-LANA in stable iSLK cells suggests that the LANA nuclear bodies form large clusters and is consistent with the observation that these clusters can interact dynamically . However , we observed that LANA nuclear bodies are transmitted faithfully to daughter cells and retain similar copy number and perinuclear distribution as observed in the parent cell . We propose that these various and heterogeneous super-molecular LANA bodies acquire an epigenetic memory allowing their stable transmission through mitotic cell division . Super-molecular structures are observed in many complex biological processes , including transcription , replication , and viral assembly . Chemical phase-changes due to the cooperative interactions of multi-valent proteins are thought to be the driving force for the assembly of these super-molecular structures [34 , 55] . Our findings support the model that modest changes in the molecular geometry of KSHV LANA can have a profound effect on the macro-molecular structure and biological function of the KSHV episome , including phenotypes in transcription , replication , and super-structures important for genome integrity and transmission ( Fig 8 ) . We propose that stable protein oligomeric architecture drives higher-order structures important for viral chromosome architecture and episome maintenance .
iSLK ( gift for Don Ganem , Novartis ) cells were cultured in Dulbecco’s modified Eagle’s medium ( DMEM ) supplemented with 10% fetal bovine serum ( heat inactivated ) and 1% penicillin-streptomycin in the presence of 1μg/ml puromycin , 250 μg/ml G418 . SLK ( gift of Don Ganem ) and BCBL1 cells ( gift of Yan Yuan ) were cultured in RPMI-1640 medium supplemented with 10% FBS and 1% P/S . iSLK RFP-LANA and iSLK GFP-RFP-LANA cells were cultured in iSLK growth medium with additional 1200 μg/ml hygromycin B . HUVEC cells ( ATCC PCS-100-010 ) grown in complete Endothelial Cell Growth Medium ( CELL applications , 211–500 ) . The KSHV bacmid BAC16 was kindly provided by Dr . Jae U . Jung and K . Brulois ( University of Southern California ) . The KSHV BAC16 clone was modified by a two-step Red-mediated mutagenesis as described previously [56] . RFP was fused in frame to the N-terminus of LANA using the pEPmRFP-in plasmid as template for PCR amplification with primers oPL6253 and oPL6254 ( S1 Table ) . A founder RFP-LANA bacmid clone was isolated and validated by restriction digest and sequencing of the entire genome . RFP-LANA bacmids were further confirmed for RFP-LANA expression and production of infectious virus . We next removed the GFP gene from the bacmid backbone of the RFP-LANA bacmid . GFP was deleted using the primers oPL7003 and oPL7004 ( S1 Table ) . The RFP-LANA bacmids ( with and without GFP ) were then used to generate the LANA oligomerization domain mutant F1037A/F1041A using the primers oPL6189 and oPL6190 ( S1 Table ) . Mutations were validated by sequencing and restriction digest to confirm integrity of the viral genome . Two clones MT1 and MT2 were further characterized . Sequencing was done on the NextSeq 500 platform using medium output run with 75bp single end . Samples bowtie2 [57] was used to align samples against the reference allowing for 2 mismatches and up to 200 multiple maps . Samtools [58] algorithm was then used call mutations and indels . Every sample was also quantified for each reference position to count number of A/T/C/G reads that had quality of at least 15 using bam-readcount tool . The sequence of Bac16 RFP-LANA WT1 , WT2 , MT1 and MT2 , not including the BAC insertion , has been submitted to GenBank ( accession no . MK143395 ) ( S1 Data ) . BAC16 and its derivatives were transfected into iSLK cells using the Effectene Transfection Reagent kit ( QIAGEN , 301425 ) . Briefly , BAC16 DNA was isolated from 2ml bacterial culture using a ZR BAC DNA Miniprep Kit ( Zymo Research ) and resuspended in 30 μl of RNase-free water . iSLK cells were seeded at 2x105 cells/well of a 6-well plate . On the next day 15 μl of BAC DNA was mixed with transfection solution and added to the cells following by transfection kit’s according to manufacturer’s protocol . On following day , transfected cells were trypsinized and transferred into 10 cm dishes and cultured in DMEM supplemented with 10% FBS and 1% penicillin-streptomycin . Two days later 1 μg/ml puromycin , 250 μg/ml G418 and 1 , 200 μg/ml hygromycin B were added to the growth media . For virus stock preparation the BAC16 stable iSLK cells were treated with 1mM sodium butyrate and 1 μg/ml Doxycycline in the absence of hygromycin B , puromycin and G418 . 4–5 days later , supernatant was collected , centrifuged ( 1 , 500 rpm for 10 min at 4°C ) and filtered ( 0 . 45 μm ) . Virus particles were pelleted trough 25% sucrose by ultracentrifugation using 27 , 000 rpm for 1 hour at 4°C . The virus pellets were resuspended in RPMI-1640 and prepared to titer the virus and future infection . 1x105 ( ~80% confluence ) cells were plated on 1cm round glass coverslips in 24-well-plate . Next day the cells were washed with PBS and fixed for 15 min with freshly prepared 2% paraformaldehyde in PBS , washed twice with PBS and permeabilized with 0 . 3% TX-100 . After washing with PBS , the cells were incubated in blocking solution ( 0 . 2% fish gelatin , 0 . 5% BSA in PBS ) for 30 min at room temperature . Primary antibody was diluted in blocking solution and applied to the coverslips for 1h followed by 3 times wash with PBS . The cells were further incubated with fluorescence-conjugated secondary antibodies for 1hr , counterstained with Dapi , and mounted in Fluoromount G medium ( SouthernBiotech ) . Images were taken at Nikon Upright Microscope using 100X objective and processed by Adobe Photoshop CS5 . Bleed-through control for RFP-LANA was performed for all IF experiments , in which coverslips were stained only with secondary antibody in the absence of primary antibody staining . We did not observe any bleed-through signals from RFP-LANA . For the α-tubulin staining of mitotic cells , iSLK RFP-LANA WT cells were seeded into 24-well plate with glass coverslips . After 72 h of growing the cells were washed with PBS , after fixed with cold 100% methanol , washed with PBS and stained with anti–α-Tubulin conjugated with AlexaFluor488 and Dapi in the end , and mounted in Fluoromount-G medium . Colocalization was determined and quantified using Nikon NIS Elements AR software , version 5 . 02 using the Spot Detection Tool to threshold foci for RFP-LANA and either Daxx , EZH2 , or ORC2 based on fluoresecence intensity , object size ( diameter ) and object shape . Binary masks were generated for each pair and the two binary layers are then combined to create a third binary layer using a having operation that identifies spatial overlap of object pixels . The binary masks were used to calculate the number of RFP-LANA overlaps with either Daxx , EZH2 , or ORC2 . High resolution , confocal images of mitotic cells in both fixed and live-cell configurations were captured using a Leica TCS SP8 WLL scanning laser confocal microscope with resonant scanner and Leica LAS-X software ( Leica Microsystems , Inc . , Buffalo Grove , IL ) , using imaging parameters which were chosen to both reduce photobleaching as well as minimize phototoxic effects . Image post-processing included importing into Huygens software for deconvolution ( Scientific Volume Imaging , Laapersveld , Hilversum , The Netherlands ) followed by maximum projection or 3D reconstruction , iso-surface application and video rendition in LAS-X . Fixed cell preparations were acquired according to Nyquist parameters using a 63X/1 . 40 oil objective , 6X zoom and a pinhole of 0 . 48 AU , and 34 z-steps through 6 . 5 um stacks , resulting in a voxel size of 51 x 51 x 200 nm . Cells were labeled with DAPI ( nuclei ) , Alexa488 ( α–Tubulin ) and RFP ( LANA ) and acquired with HyD detectors in sequence to maximize signal and minimize cross-talk . 6 line accumulations allowed laser intensities to be kept to 0 . 5% at 405 nm , 0 . 5% at 488 nm and 2% at 557 nm . Original time-lapse sequences were acquired with cells incubated in 35 mm MatTek dishes ( MatTek Corp . , Ashland , MA ) and maintained in a Tokai-Hit stage-top incubation chamber ( Tokai Hit , Shizuoka-ken , Japan ) at 37 °C and 5% CO2 . Imaging hardware parameters were adjusted to preserve cell viability , including laser intensity settings of 0 . 3% at 488 nm for GFP and 2 . 0% at 555 nm for RFP , 4 line accumulations and use of HyD detectors . Stacks of images were captured with a 40X/1 . 30 Oil objective and 2X zoom at 10 locations over 48 hrs with a 10 min sampling interval , totaling 289 time points at each location . Each stack was 26 um thick and comprised of 34 sections in 0 . 8 um steps . The following antibodies were used for immunofluorescence studies: rabbit anti-Daxx ( Sigma , D7810 ) , rabbit anti-ATRX H-300 ( Santa Cruz , Sc15408 ) , rabbit anti-PML ( Bethyl , A301167A ) , rabbit anti-EZH2 ( Cell Signaling , 4905S ) , rabbit anti-H3K27me3 ( Active motif , 39155 ) , mouse anti-ORC2 ( MBL , M0553 ) , and mouse anti-αTubulin/Alexa488 ( Invitrogen , 322588 ) . Secondary antibodies AlexaFluor488 or AlexaFluor594 were purchased from Invitrogen . The following antibodies were used for Western blotting: mouse anti-ORF50 ( provided by Erle Robertson , UPENN ) , mouse anti-ORF45 ( provided by Yan Yuan , UPENN ) , rat anti-LANA ( Advanced Biotechnologies Inc . , 13210 ) , rabbit anti-Daxx ( Sigma ) , and anti-actin-HRP ( Sigma , A23852 ) . Antibodies used in ChIP assay include: rabbit polyclonal antibodies to histone H3K4me3 ( Millipore , 07473 ) , histone H3K27me3 ( Active motif , 391155 ) , total histone H3 ( Bethyl ) , ORC2 ( MBL , M0553 ) , Rad21 ( Abcam , ab992 ) , CTCF ( Millipore , 07729 ) , or rabbit IgG ( Santa Cruz Biotechnology , sc-2027 ) , and rat polyclonal anti-LANA ( Advanced Biotechnologies Inc . ) . Equal amounts of protein extract in RIPA buffer ( 50mM Tris-HCl ( pH8 . 0 ) ; 150mM NaCl; 1% NP-40; 0 . 5% Sodium deoxycholate; 0 . 1% SDS; 1mM EDTA ) were resolved in 8–16% Novex Tris-Glycine gels ( Invitrogen ) , and then transferred onto a PVDF membrane ( Millipore ) , where they were incubated with specific antibodies followed by HRP-conjugated secondary antibodies ( BioRad ) and ECL reagents ( Millipore ) for detection . RFP-LANA WT and MT iSLK cells were washed 2 times with cold PBS , and then lysed in cold lysis buffer ( 20 mM Tris , pH 8 . 0 , 137 mM KCl , 1 mM EDTA , 1 . 5 mM MgCl2 , 10% Glycerol , and 1% Triton X-100 supplemented with 1 mM DTT and 0 . 1% mammalian protease inhibitor cocktail mix ) for 30 mins on ice . Cell lysates were centrifuged at 13000 rpm for 10 mins , and the supernatants were precleared with Protein G Sepharose beads ( GE Healthcare ) for 60 mins at 4 °C with rotation . One ml of precleared lysates ( ~5 x 106 cells ) were immunoprecipitated with either rat anti-LANA ( Advanced Biotechnologies Inc . ) or mouse monoclonal anti-p53 ( CalBiochem ) or rabbit anti-DAXX ( Sigma ) overnight at 4 °C with rotation . The immuno-complex was collected with Protein G sepharose beads with rotating at 4 °C for 3 hrs , and the beads were washed 3 times with BC300 ( 300 mM KCl , 20 mM Tris-HCl , pH 8 . 0 , 0 . 2 mM EDTA , 10% glycerol , and 10 mM β-mercaptoethanol ) followed by once with BC100 at 4 °C . Pulled down proteins were eluted by boiling with 2x Laemmli buffer ( 100 mM Tris-HCl , pH 6 . 8 , 4% SDS , 0 . 2% Bromophenol Blue , and 20% Glycerol ) , and were subject to SDS-PAGE and Western blot analysis . The amount of intracellular KSHV DNA and bacmid DNA were determined by quantitative PCR ( qPCR ) analysis using primers specific for KSHV genome ( listed in S2 Table ) . The data were normalized to the hygromycin DNA region , as described previously [32] . HEK293T cells were seeded at a density of ~200 , 000 cells/well in 12-well plates . The following day , the cells were transfected with a Gaussia luciferase reporter plasmid ( containing a single LBS1 or LBS2-LBS1-LBS3 ) and a plasmid for expression of a fusion protein containing an N-terminal FLAG tag , the VP16 ( activation domain ) , and the wild-type ( wt ) or mutant ( F1037A , F1041A ) LANA DNA-binding domain ( DBD ) . An empty FLAG vector was used to ensure that the total amount of DNA per transfection was equal to 225 ng . Transfections were performed using Lipofectamine 2000 at a ratio of 1:3 ( mass of DNA per volume of transfection reagent ) in antibiotic-free media . After 6 hours , the antibiotic-free media was replaced with media containing antibiotics . Approximately 18 . 5 hours later , luciferase measurements were made by transferring 40 μL of media and 10 μL of substrate ( BioLux Gaussia Luciferase Assay Kit , New England Biolabs ) to a 96-well plate and measuring luminescense with an Envision plate reader ( PerkinElmer ) . Cells were lysed with RIPA buffer containing 1 mM PMSF and supernatants were run on a NuPAGE 4–12% Bis-Tris Gel ( Invitrogen ) and transferred to a Power Blotter nitrocellulose membrane ( Invitrogen ) using a Power Blotter ( Invitrogen ) . The membrane was probed using anti-FLAG M2-Peroxidase antibody ( 1:10 , 000 ) ( Sigma-Aldrich ) or β-tubulin loading control antibody ( 1:4 , 000 ) ( Fisher ) . ChIP assays were performed as described previously [59] . Briefly , cells ( ~ 1 x 107 ) were crosslinked in 1% formaldehyde with shaking for 15 min , quenched by the addition of glycine to a final concentration of 0 . 125 M , and lysed in 1 ml SDS lysis buffer ( 1% SDS , 10 mM EDTA , and 50 mM Tris-HCl , pH 8 . 0 ) supplemented with 1 mM PMSF and protease inhibitor cocktails ( Sigma-Aldrich ) . The lysates were sonicated with a Diagenode Bioruptor , cleared by centrifugation to remove insoluble materials , and diluted 10 fold into IP Buffer ( 0 . 01% SDS , 1 . 1% Triton X-100 , 1 . 2mM EDTA , 16 . 7mM Tris pH 8 . 1 , 167mM NaCl , 1 mM PMSF , and protease inhibitors cocktail ) for IP reaction at 4°C overnight . Each immune complex was washed five times ( 1 ml wash , 10 mins each ) in ChIP related wash buffer at 4°C , eluted by addition of 150 μl Elution buffer ( 10mM Tris , pH 8 . 0 , 5mM EDTA , and 1% SDS ) at 65°C for 30 min , and the elutes were placed at 65°C for overnight to reverse cross-linking . The elutes was further treated with Proteinase K in a final concentration of 100 μg/ml at 50°C for 2 hrs , and ChIP DNA was purified by Quick PCR Purification Kit ( Life Technologies ) following the manufacturer’s instruction . ChIP DNA was assayed by qPCR using primers specific for indicated KSHV regions and quantified as % input . 3C experiments were carried out as described previously [33] , with minor modifications . PCR data were normalized to cellular actin . Briefly , 5×106 KSHV iSLK WT or MT cells were cross-linked with 1% formaldehyde at room temperature for 10 min , followed by the addition of glycine at the final concentration of 125 mM for 10 min at room temperature . Cellular pellets were washed once with cold PBS and resuspended in 250 μL of ice cold lysis buffer ( 10 mM Tris-HCl , pH 8; 10 mM NaCl; 0 . 2% Igepal CA630;1 × complete protease inhibitor; 11836145001 Roche ) for 30 min with rotation at 4°C , centrifuged for 5 min at 2500 g at 4°C and then washed once with 500 μL of ice-cold lysis buffer . Resultant pellets were gently resuspended in 100 μL of 0 . 5% SDS and incubated at 65°C for 10 min . After addition of 50 μL of 10% Triton X-100 to quench the SDS and 290 μL of distilled water , samples were incubated at 37°C for 15 min , centrifuged for 5 min at 2500 g at 4°C and washed once with 500 μL of ice-cold lysis buffer . Pellets were resuspended in 200 μL of 1X restriction enzyme buffer and digested with 100 units of BamHI overnight at 37°C . To inactivate the restriction enzyme , the digested samples were incubated at 65°C for 20 min . The samples were ligated in 1 mL of ligation buffer ( 1X T4 ligase buffer , 1% Triton X-100 , 8μg/mL BSA , 10000 CEU of T4 DNA ligase ) at room temperature for 6 hours with slow rotation . Ligated samples were centrifuged for 5 min at 2500 g at 4°C , resuspended in 368 μL of lysis buffer and treated with 10 μl of 20 mg/ml Proteinase K , 20 μL of 10% SDS , 40 μL 5M sodium chloride at 65°C overnight to reverse cross-linking . Genomic DNA was extracted with phenol/chloroform , resuspended with 250 μl of distilled water and subjected to qPCR using primers listed in S3 Table . | KSHV genomes persist in large nuclear bodies in latently infected cells . The KSHV encoded nuclear antigen LANA is required for the efficient replication and stable maintenance of viral genomes during latent infection . LANA is also known to form oligomeric structures , but it is not known how these structures contribute to LANA function in living cells . Here , we show that LANA oligomerization is required for cooperative binding to the KSHV terminal repeat ( TR ) , and the recruitment of the Origin Recognition Complex ( ORC ) to viral TR . LANA oligomerization is required for a chromosome conformation DNA loop between TR and the LANA promoter implicated in LANA transcription autoregulation . LANA oligomerization is also required for formation of large nuclear bodies that colocalize with DAXX , EZH2 , ORC2 , but not PML . LANA nuclear bodies distribute along the nuclear periphery , and their arrangement is transmitted faithfully to daughter cells during mitotic cell division . Finally , we show that KSHV genomes containing mutations in the LANA oligomerization interface fail to maintain the complete viral genome , suggesting they are defective in DNA replication or repair . These findings reveal new mechanisms of LANA episome maintenance through formation of higher-order chromosome-conformations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"cell",
"cycle",
"and",
"cell",
"division",
"pathogens",
"cell",
"processes",
"microbiology",
"dna-binding",
"proteins",
"viral",
"structure",
"viruses",
"viral",
"genome",
"dna",
... | 2019 | LANA oligomeric architecture is essential for KSHV nuclear body formation and viral genome maintenance during latency |
Over the last decade , implementation research and a science of global health delivery have emerged as important vehicles to improve the effectiveness of interventions . Efforts to control neglected tropical diseases ( NTD ) operate in challenging circumstances and with marginalized populations , making attention to context-specific details particularly relevant . Socio-anthropological insights have much to offer a science of NTD delivery . In this paper , an accessible and actionable framework for understanding NTD intervention effectiveness , based on socio-anthropological research , is presented and its utility for program planning and monitoring and evaluation is outlined . The framework was developed inductively by comparatively analyzing three rapid ethnographic studies undertaken in Eastern Africa ( 2010–2013 ) on three different large-scale NTD interventions: rabies elimination in Tanzania , sleeping sickness control in Uganda and the prevention of parasitic worms in Zambia . The framework includes five “intervention domains” where the effectiveness of these interventions was negotiated and determined at the local level . This involves: 1 ) the terrain of intervention ( including seasonality and geographical variability ) ; 2 ) community agency ( including local knowledge , risk perceptions , behaviors , leadership and social pressure ) ; 3 ) the strategies and incentives of field staff ( skills , motivations , capabilities and support ) ; 4 ) the socio-materiality of technology ( characteristics of intervention tools and the adoption process itself ) ; and 5 ) the governance of interventions ( policy narratives , available expertise , bureaucracy , politics and the utilization of knowledge ) . The paper illustrates the importance of each of these domains by drawing on the case study research , presenting lessons learnt and practical recommendations for how such insights could improve intervention delivery . To help close the gap between efficacy and effectiveness in NTD programs , it is important that field staff: 1 ) generate meaningful knowledge about contextual factors; 2 ) use this knowledge to tailor field strategies; and 3 ) create routine mechanisms to account for the dynamic process of implementation itself . The framework presented here offers a simple analytical tool to strengthen these knowledge-to-action relationships existing project planning tools , drawing on the insights of socio-anthropology .
It has long been argued that social science perspectives have a great deal to offer the world of global public health . While strides have certainly been made , with the integration of socio-anthropologists and others in research and programs now more mainstream , progress is still slow and uneven [1–3] . This is especially so in countries where neglected tropical diseases ( NTDs ) are most common . Here , in contexts of poverty and affliction , health system research tends to be down the list of priorities and disciplinary divisions remain more firmly entrenched . At the same time , wealthier scientific and development partners are often focused more on generating evidence for new tools and technologies . The question of how to bring existing interventions to scale , embed them successfully in health systems and ensure they reach their full potential in diverse local settings , across hundreds of millions of people globally , remains somewhat of a twilight zone . In the era of Sustainable Development Goals ( SDGs ) , implementation research has ascended to the top of the priority list in global health , in light of the emergence of “implementation science” and efforts to create a “science of global health delivery . ” While there are a variety of definitions , Allotey et al . [4] defined implementation research as: Against this backdrop , different frameworks are calling attention to how social context and social interaction influence global health implementation . Damschroder et al . [5] proposed a meta-theoretical framework of different interacting domains including: how the intervention fits into implementing organizations; the external social and economic environment; the perceptions , abilities and motivations of planners and implementers; and the process of implementation itself: planning , engaging , executing , reflecting and evaluating . Gruen et al . [6] , drawing on participatory action research , proposed that health programs be viewed as complex social ecosystems where diverse stakeholders interact with very different norms and interests , and where constant engagement , reflection , research and adaptive learning are needed to keep them on track . Frost and Reich [7] proposed what they called the “access framework” , focused on how acceptability , affordability , availability and organizational architecture shape the adoption process for health technologies in resource-poor countries . Importantly , Obrist et al . [8] added a focus on livelihood vulnerability and social resilience . This ethos has also influenced the NTD community; see , for example , Implementation Research for the Control of Infectious Diseases of Poverty [9] and The Global Report for Research on Infectious Diseases of Poverty [10] . Social scientists have provided important inputs into this agenda , and a new “trans-disciplinary vision” for how to conceptualize the control of diseases of poverty , especially those with more complex disease ecologies , has emerged [11] . According to Prentice [17] , ethnographic research in the field of global health has four main principles: it uses fieldwork to build theory , it emphasizes meaning and classification , it explores the negotiated nature of reality , and it emphasizes the central role of context . It can challenge our view of the world and our place in it . Social theories remind us that interventions are dynamic , can have unintended consequences , are socially constructed , involve power dynamics and are sites of negotiation , even contestation , between different social groups [18] . To the ethnographer , interventions are not clean , neutral activities but are complex and messy: a social arena where histories , politics and social conflicts are inevitable . Anthropological approaches in global health have evolved over the last few decades in response to the push-and-pull of funding streams , scientific networks , methodological innovations and the priorities of program planners and managers , among other factors . These tend to emphasize more rapid methodologies , focused on expedient and actionable forms of knowledge orientated around specific operational questions . This has underpinned the growth of a class of rapid anthropological studies in the 1990s , for example , in WHO-supported childhood diarrheal , malaria , HIV and other programming [12–14] . Studies on NTDs have evolved in parallel to include substantial work on illness categories , drug use patterns , community participation , gender dimensions and community perceptions and responses to interventions [15] . Not all of these , of course , are based on rapid approaches ( which have acknowledged pitfalls and risks [13] ) , and some also draw on action research methodologies , such as participatory rural appraisal ( PRA ) , that use community mapping exercises , diagrams and flowcharts alongside more traditional qualitative methods and participant observation [16] , borrowing from the field of international development and humanitarian emergency . One of the challenges for socio-anthropologists is to bring a deeply contextualized knowledge into the planning and implementation process , in ways that are actionable but not reductionist , taking account of the complexities involved , both from a methodological and social standpoint [19–20] . Monitoring and evaluation ( M&E ) frameworks and approaches have borrowed from qualitative methods , and many field staff have been trained in these , using them in routine program planning activities . But much of this applied knowledge is based on overly simplistic tools , like the knowledge , attitude and practice ( KAP ) survey [19]; more flexible and contextualized approaches , like systematic comparative ethnography [20] are a welcomed addition . Despite more biomedical scientists and public health experts recognizing the benefits of more flexible anthropological insights [2] , many global health programs still focus predominately on quantitative metrics and struggle with how to conduct , report and operationalize qualitative and ethnographic data . The aim of this article is to outline a framework for socio-anthropological insights into intervention effectiveness that could , theoretically , assist in orientating M&E and/or an operational research agenda . The framework was developed inductively by analyzing the results of rapid ethnographic studies conducted on three intervention case studies in Eastern Africa ( sleeping sickness , rabies and parasitic worms ) from 2010–2013 . This work focused on understanding intervention effectiveness through issues of coverage , adoption , participation and use of health technologies , and drew on the disciplines of medical anthropology , sociology , science and technology studies , development studies , communication studies and public health .
The three case studies , on which the framework is based , involved three very different health interventions , with the hope that this variability would provide unique insights into how NTD interventions are negotiated at the local level and the various areas where effectiveness is determined . Following an ethnographic approach , the fieldwork relied on mixed methods , combining quantitative data on coverage , uptake and use of health technologies with substantial and in-depth qualitative research , participant observation , document review and ethnographic notes . For a description of the methods see [21] , as well as individual peer-reviewed publications in PLOS NTD [22] , Medical Anthropology [23] and Geoforum [24] . For greater detail on the case studies , see these individual publications . There were five major similarities common to these three interventions that provide for strong comparisons and insights . First , they were all “mass interventions” covering large geographical areas and socio-economic contexts that were financially supported by international donors and planned by technical experts . These projects represented “scaled-up” NTD interventions , aimed at targeting hundreds of thousands of farmers and cattle ( Uganda ) , tens of thousands of dogs ( Tanzania ) and many hundreds of villages ( Zambia ) . Second , the projects had bold targets that aimed for big impact–as noted in their names and goals , they aimed to “eliminate rabies” , “stamp out sleeping sickness” and achieve “total sanitation . ” The assumption was that these interventions would showcase the cost-effectiveness and feasibility of preventing NTDs in rural Africa . Third , they relied on local participation , behavior change and the adoption of prevention practices and technologies , such as latrines ( Zambia ) , veterinary insecticides ( Uganda ) and dog vaccines ( Tanzania ) . These were , in turn , driven by specific justifications that framed this technology as “appropriate” for rural African contexts: rabies vaccination in Tanzania was free and has very minimal side-effects on dogs; community-led total sanitation ( CLTS ) is an innovative WASH approach , deemed superior to past subsidy-based sanitation approaches in Zambia because it promised to be “community-led” and reliant on locally appropriate building materials; the Stamp Out Sleeping Sickness ( SOS ) public-private partnership in Uganda aimed to link agro-veterinary business to sleeping sickness control and livestock improvement by creating new systems of veterinary drug delivery that built on existing practices . These were all “low-cost” and “low-tech” health technologies , underpinning the hope that meeting project targets could be achieved within a short period of time with relatively modest funding . Fourth , the delivery of the interventions were done by district and sub-district actors: government staff , local leaders , extension workers , volunteers and private businessmen . Lastly , specific incentive structures were used to mobilize these people and to motivate them to deliver the intervention and to conduct social mobilization , community engagement and risk communication , with the idea that such approaches would be “locally-led” and “sustainable . ” As with any comparison , there were also important differences . The three interventions involved different types of pathogens–sleeping sickness , rabies and parasitic worms–in three different Eastern African countries with diverse cultures , landscapes , languages , politics and other contextual factors . Similarly , the three interventions all used very different approaches ( top-down , participatory and market-driven ) , institutional arrangements ( WHO country office , district teams and a public-private partnership ) , control technologies ( vaccination , social mobilization for pit latrines and veterinary insecticides ) as well as delivery networks and local incentive structures to enroll support and participation . These are all summarized in Table 1 . Unexpectedly , the individual coverage data found in my large-scale surveys ( and validated by my qualitative and observational data ) showed that all three of the interventions achieved disappointingly low uptake: 25% vaccination coverage in Tanzania ( from a survey of n = 6 , 157 households ) , 31% latrine coverage in Zambia ( n = 922 households ) , and an 8 . 7% market share of the SOS-supported insecticide ( known as Vectocid ) in Uganda ( survey of n = 87 veterinary shops ) [22–24] . These projects not only shared these low coverage rates; many of the most salient reasons for why dogs were not vaccinated , latrines were not constructed , and veterinary insecticides were not purchased and used by livestock keepers had many underlining commonalities . This low coverage was also not inevitable . There were a number of adaptive pathways–as the case studies all make clear below and in the individual case studies [see 22–24]–that could have been used to increase coverage to more acceptable levels both initially and as the interventions progressed over time . Many of these required only modest changes in operational plans , suggesting that the research approach taken here could have substantially improved intervention delivery and population health . In this paper , a framework for socio-anthropological engagement in NTD intervention effectiveness research and program planning is outlined that draws on a synthesis and analysis of these three case studies . This framework seeks to convey the key areas where effectiveness is negotiated as a complex set of interactions between ecosystems , local communities , animals , implementers , health systems and policymakers , within their broader biosocial context . As shown in Fig 1 , this has five “effectiveness domains . ” Of course , the framework does not intend to be fully comprehensive; rather , it is presented here as a flexible conceptual tool to assist those planning and implementing NTD control to think about these critical issues . While these three interventions provided the initial analytical lens for the framework , it is worth noting that my thinking has also been informed by subsequent research and control programs , published on mosquito-borne diseases ( lymphatic filariasis , malaria and Zika ) in Haiti [25] , cysticercosis and helminths in Lao PDR [26] , rabies in Indonesia [27] , cystic echinococcosis in Morocco [28] and other NTDs [29–30] . Overall , this breadth of experience ( of different diseases and contexts ) has only strengthened my conviction of the utility of such a framework in assisting to inform NTD control interventions .
Global health interventions function over a “local space” where technologies and tools are deployed and issues of coverage and impact are measured and determined . These terrains of intervention are created through strategic decisions based on available resources , political expectations and epidemiological knowledge , and aligned with regional , ecological or district boundaries . There is a specified , and unfortunately often all-too-short , time period where the “field” needs to be understood and transformed [31] . Outside the usual realm of “socio-cultural practices” , diverse social groups , governance histories and micro-ecologies exist within these geographies; local livelihoods interact with land-use patterns , seasonal fluctuations , human movement and migration and new socio-economic pressures , many times outside expectation [32] . My research showed that this variability is important but can easily be subsumed by the challenges of creating technical delivery systems . First , seasonal and geographical variations in local livelihood systems can be in direct conflict with intervention delivery schedules . Low levels of attendance at CLTS sanitation empowerment meetings in Zambia were due to the program starting implementation when farmers were busy harvesting their crops . Seasonal changes in crop farming and livestock management drove the migration of pastoralists in the Tanzanian rabies case study , the social group with the largest number of dogs . However , the elimination program did not account for this–it was one of the most important reasons for the low vaccination rate . The full-scale ( and military/police-led ) eviction of thousands of livestock-keepers from the Kilombero Valley after my fieldwork , due to concerns about soil erosion and land carrying capacity , demonstrated just how important population movements can be . Planning effective dog vaccination in subsequent years would require accounting for these communities in the wider WHO elimination area . See [33] for a review of this issue in relation to NTDs more generally . These dynamics clearly have a major impact on intervention coverage , showing that there is often an optimal “window” for interventions in a given geography , orientated around these seasonal changes and the livelihood patterns of high-risk groups . Geographies need to be problematized as diverse and coupled to livelihood-seasonal change instead of the predominate tendency to conceptualize them as uniform and singular . Seasonality also influences the purchasing power of households . In the Ugandan sleeping sickness case study , reductions in income during the dry season meant that there was little money available to buy insecticides . This happened to correspond with seasonal tick and tsetse population reductions , mitigating an otherwise dramatic influence of seasonality . But it also meant that the veterinary drug shops established by the SOS program struggled , making it harder for these professionals to make a living . The fragility of local economies and livelihoods in the Ugandan study area were also dramatically affected by a foot-and-mouth disease ( FMD ) quarantine , an epidemic of cassava mosaic disease , severe flooding followed by drought and a longer-term crisis of land fragmentation , due to high population growth . These vulnerabilities led to a volatile market for the veterinary drug sellers and their community-based animal health workers . Another , perhaps obvious finding , is that intervention staff have a harder time moving along local terrains during the wet season . The large distances , with poor road conditions , created high translation costs for the SOS veterinarians in Uganda , reducing the time they were willing to educate farmers , at least when direct financial support for community-based education was withdrawn by the financing partners . Large distances also dissuaded the animal health workers from spraying cattle–they did not like “running around chasing people” ( as they would say ) –one reason , among others , for why they preferred the more profitable work in injectable ( treatment ) drugs . Seasonal variation and human movement patterns had an impact on local disease risk perceptions and influenced people’s willingness to adopt health technologies . For example , in Zambia , the fact that people defecated more in their agricultural fields than in their latrines ( or around their village ) in the rainy season , helped to normalize open defecation and de-motivate households to maintain latrines in the village since construction could only address “half our sanitation problem . ” Seasonal weather and changing ecology contributed to many latrines collapsing . Villages that were densely settled , due to historic land-use drivers going back to British colonialism , had created bylaws preventing some households from building latrines due to concerns about contamination and miasmic notions of disease spread . Generally , the distribution of disease is seldom uniform within a certain intervention area , but often clustered in “hotspots . ” Interventions have to make strategic spatial decisions regarding delivery and coverage: the placements of central vaccination point in Tanzania , the location of veterinary drugs shops in Uganda and which villages will be triggered for CLTS in Zambia . In the three case studies , delivery was always conceptualized , by planners , at an abstract district-level without adequately considering the social ecology or epidemiology of the area . In Zambia , certain conditions–many of these tied to ecological characteristics ( i . e . access to latrine material ) and levels of socio-economic development–offered CLTS the best chance of having an impact on sanitation . In Tanzania , the majority of the dog population was found in very remote pastoralist villages , despite the fact that these were not very well covered by the vaccination program . In Uganda , the SOS supported veterinarians did not focus their efforts on villages with active sleeping sickness patients , despite hospital staff ( and hospital records ) being available in the local treatment hospital ( the focus was on meeting monthly sales targets and no shop was actually located in the most highly-endemic sub-county ) . Just as epidemiologists now speak about “super-spreaders” [34] , intervention planners should operate according to targeted strategies that focus on high-priority areas . These examples all show the importance of diversifying our idea of the “intervention terrain” as we seek to use knowledge to improve implementation , paying attention to local livelihoods , human movement , seasonal change , geographic variability and epidemiological hotspots . As with the terrain of intervention , there is also a need to engage with social difference and consider variations in human perspectives , attitudes , logics , social organization , interests and agency . There is a tendency to reify communities as the ultimate target for interventions [35]; but a community is a social construct , a network of relationships and dependencies , and not a geographically bounded unit . Rather , intervention terrains have diverse social groups with differences in wealth , ethnicity , livelihoods , power , knowledge , cultural norms and needs , capacities and constraints . Some people will be more interested than others in the intervention , and more able to adopt technologies and health behaviors . Some will refuse the intervention and may influence others to do the same . Some will recommend changes . This means that planners and implementers need to understand the “public” in public health: they are able to exert influence , make their voices heard and transform plans and policies [36] . But they also have constraints: “throughout the world , those least likely to comply are those least able to comply” [37] . My research found that the intervention adoption process , for insecticides , vaccines and pit latrines , reflected the “uneven playing field” [38] of the rural East African village and deep historical trends of socio-economic and political marginalization . Interventions to alleviate diseases of poverty need to navigate these contexts of destitution , deep inequality and even squalor . The reality is that adoption of health technologies and prevention practices tend to align with higher levels of material wealth and social capital at local level . In Zambia , for example , latrine ownership acted as a symbol of modernity , and it was those households with better education , mobility , access to government subsidies , housing , food security and social networks who build and used them ( before and after CLTS ) . Decades of war , rapid inflation , population growth and other developmental challenges were repeatedly stressed in my Ugandan research to explain why people ( sometimes called the “disorganized people” ) preferred the cheaper non-tsetse effective insecticides ( those only effective on ticks and not the tsetse flies that spread deadly sleeping sickness ) , despite many knowing that they were inferior products . In Tanzania , canine vaccination was not necessarily influenced by wealth but rather by the motivation for people to keep dogs–dogs that were better cared for and had a defined role in the household were more likely to be vaccinated . Owners were also more able to bring these dogs to the central vaccination point because they were better-behaved dogs . This shows that certain social groups and geographical locations are more likely to be receptive to the intervention . Motivation to comply with the three interventions , and adopt prevention practices , had a great deal to do with how local people understood the benefits involved . Exposure to “scientific” frames of reference facilitated by veterinary and health extension workers , ideas of social and communal responsibility and a desire to be “modern” were found to play important psychological and cultural roles in all three case studies . Generally , abstract biomedically-defined explanations only went so far in persuading people; rather , what was directly observed , experienced and narrated in locally-understood ways by the community had much more persuasive power . For example in Uganda , many farmers ( who were used to spraying insecticides directly onto ticks to kill them ) found it hard to understand that cattle sprayed with insecticide could kill a tsetse fly who came to feed on it many hours later . A “hybridism” between experience , local knowledge of disease and biomedical information predominated , something that risk communication and education , which was only done sporadically in these interventions , could have better incorporated . Local ideas about the value and usefulness of practicing disease prevention strategies were important to people’s motivation , and based on personal experience . In Tanzania , the fear of rabies was frequently mentioned as a motivational force , since people had heard stories ( often through their social networks ) of previous victims and also feared mass dog culling of non-vaccinated dogs ( which local government sometimes implemented on its own accord ) . The level of input ( in time and resources ) that people had to invest to rabies control ( annual vaccination ) was minimal compared to the other case studies , reflecting the lower end of the “participation scale”–as discussed by Rifkin [39] . This certainly helped amplify people’s willingness to comply . In Uganda , experiences of tsetse and ticks and a villages’ location relative to swamps and bushes ( breeding areas ) were important causes for the level of priority given to sleeping sickness and cattle diseases . Having an animal die from a tick-borne disease was the most often mentioned reason for why farmers began to pay and use insecticides on a regular basis . De-motivational forces were found to be significant in Zambia and Uganda , where adopting long-term prevention practices were framed as a “gamble” and “risk” , partially because it was never guaranteed to work . This is part of the challenge with adopting health promotion behaviors: they are evaluated based on costs , effort , potential benefits , and in relation to the other multiple vulnerabilities people face . Many simply concluded that the control of neglected diseases was not a priority for them . These were “rare” diseases , and it was better for people to invest their time into something else . This was especially the case when transmission pathways involved more than one route , such as for sanitation-related diseases where one could still get sick despite having a latrine , for example . An important force behind participation and the promotion of community compliance was the exercise of power , influence and authority at the local level . Punitive efforts included “bylaws” to lock latrine doors and impose “fines” for not having a latrine in Zambia . In Tanzania , this included “village laws” to cull all non-vaccination dogs , dog registration and financial reciprocation laws if a rabid dog attacked someone . Social pressures and simplified local narratives about why people should comply with the interventions were important in normalizing participation and making the case , in locally understandable terms , for compliance and involvement . These included: ticks can kill your cattle; open defecation makes people in the village sick; and not vaccinating your dog can kill your neighbor . These did not necessarily focus on specific diseases , and tended to use stories about local people’s experiences . There was also a different angle to public agency; a household would build a latrine to show their wealth and prestige , to further develop their building techniques and assist in mobilizing ( or rebuking ) other community members . In this case , those who did not act on these activities were frequently looked down upon , and sometimes mocked and ridiculed as “unsanitary citizens” , even if the ultimate reason was their extreme poverty . Briggs and Mantini-Briggs [40] describe this process in their superb ethnography of the 1990s cholera epidemic among indigenous communities in Venezuela . Alternatively , all interventions depended on the support and legitimizing labor of local leaders , who were essential in mobilizing community members to attend meetings , spread information and manage aspects of the intervention . In Tanzania , village leadership arranged the location of the central points and disseminated information about the campaign , often door-to-door . In Zambia , poor leadership–where many village leaders themselves , for example , did not have a latrine–contributed to a lack of cohesion and motivation that was only overcome in a few villages with active youth groups , income generation groups and a stronger women leadership culture . A similar case was found in Uganda , where the business aspect of the intervention meant that these leaders only mobilized the community if they were paid a small “motivational” fee . Agency , of course , is exerted not only by the recipients of interventions but also by those charged with implementation [25 , 31 , 38 , 40] . The context of the health system , including aspects like human resources , information systems , drug supply chains , basic infrastructure and the culture of management and care , are incredibly important . Interventions require enrolling the support of actors embedded within these systems , and it is often these health staff , community outreach workers and volunteers that are responsible for actual implementation , translating between the different logics and interests of the intervention and the community [29] . In this process , the interests and relationships of these field staff play an overwhelming role in success or failure . In the case studies , the delivery and planning of interventions was the task of district bureaucrats , extension workers , local leaders , private shops and volunteers . Outside of the training room and models put forth by outside experts , it was these stakeholders ( and their working norms , cultural knowledge , incentives and motivations ) that shaped the course of events through field-level decisions and social encounters . In Zambia , for example , many of the local champions trained by the CLTS project used the approach in a piecemeal fashion: they did not use the word “shit” ( which is used to provocatively illustrate the fecal-oral-disease route ) and rarely conducted any follow-up visits with village groups . But they were also clearly not very competent: many were not very well respected by these communities , lacked technical knowledge about latrine construction itself and could not record data properly . In response , people were reluctant to attend community meetings and voiced expectations for hardware subsidies ( instead of following the community-driven ideology of CLTS ) . While local environmental health technicians ( EHTs ) , employed by the Zambian Ministry of Health , were to oversee these volunteers , they were never given any clear expectations , resources or guidance in how to do so . These EHTs themselves operated under severe staff shortages ( I often observed clinic cleaners acting as doctors and nurses at these local health centres ) , corruption , inadequate supplies and low morale . In Tanzania , veterinary workers were tasked with selecting the central point and mobilizing dog-keepers ( in collaboration with village leaders ) , ensuring adequate supplies of vaccines and administering the vaccines . Although they often enrolled the support of local leaders , I found that many of the vaccine central points were placed in locations close to major roads , far away from the more remote ( and difficult to access ) areas where , unfortunately , there were more dogs . Information dissemination to these rural locations was limited . As Pigg [41] has argued in relation to traditional birth attendants in Nepal , the tendency for projects to categorize local actors involved in implementation based on overarching stereotypes can create simplistic assumptions , leading to inappropriate delivery structures . Staff may need to be fired and replaced , making staff quality control and regular team meetings an important component of an effective intervention system . Many of the volunteers in Zambia had been selected by local political allies with the expectation of gaining monetarily from the program , without doing any serious work . They were provided bicycles and promised money for achieving a set number of Open Defecation Free ( ODF ) villages but had little oversight , provision of material ( pencils , paper and airtime ) and clearly defined benefits rewarding hard work . Motivation and support offered to volunteers has also been noted in the ongoing delivery of ivermectin for the control of onchocerciasis in West Africa , where high dropout rates have revolved around a lack of incentives and supervision , long travel distances , other livelihood duties , drug supply problems and working in areas not familiar to the volunteers–volunteers have also been noted to perform better where money has been provided , such as in polio vaccination [42–43] . Field staff have expectations that need to be met to ensure their continued enrolment and performance . For them , global health interventions are a source of work , of salary , prestige and livelihood–as described by Geissler [31] in his ethnographic work in Kenya . The insecticide sprayers supported by the SOS vet shops in Uganda , for example , expected regular workshops , training , subsidies , drugs on credit and various types of free materials such as spray pumps , overalls and gumboots . But these were not provided , at least as often as they assumed would be the case , which reduced their commitment to project goals . Local leaders in Tanzania expected some small financial incentives for mobilizing farmers and livestock keepers in remote areas . In some cases , these were not provided , and coverage in these areas was considered lower than others . When we speak about interventions , we are also talking about technology . The ways that technology characteristics and features embed , and are embedded within , social relationships offers an interesting conceptual approach to explore how global health interventions are delivered . We can call this approach socio-materiality or the “social life” of technology [44–46] . The three case studies showed how the technologies themselves mediated adoption , delivery and use patterns . In Tanzania , rabies vaccines required cold-chain storage that needed to be delivered at specific central points and whose supply depended on international procurement and adequate syringes . Preferences for non-tsetse insecticides in Uganda involved the characteristics of the insecticides and the effect of the drug: the smell , color , packaging , residual period and mode of action . The fact that many farmers only used insecticides to target tick predilection sites was responsible for the continued preference for amitraz products , which are not effective on tsetse flies , only ticks . To use the insecticides , farmers had to buy or borrow spray equipment ( which often broke ) and needed protective gear . Without them , they used water bottles , which meant that dilution rates and application methods were not ideal . A different scenario presented itself in Zambia . Weather and insects destroyed poorly constructed latrines while the level of faecal matter in the pit created impressions that they were “unhygienic . ” People had to negotiate the landscape as they searched for the more durable logs to construct the latrine base and acquire the necessary material ( bricks or bamboo ) to make the superstructure; these were increasingly difficult to find in some areas due to land-use pressures and environmental change . Latrine construction was variable , with many different designs . Building was influenced by the technical knowledge of the owner and builder and their relationships . Latrines required maintenance , had different longevities and were spaced away from homes . Access to durable materials and building techniques played a major role in influencing latrine construction , use and maintenance . Technologies , social forms and ecological characteristics are embedded within a dynamic web of causality . Viewing health technologies as having a socio-material existence promotes understanding these dynamics as an essential step in intervention effectiveness . A true science of global health delivery cannot simply focus on the deployment of predetermined interventions but should also critically evaluate the appropriateness of these interventions , their policy models and wider political economy . Negotiating bureaucratic procedures , knowledge flows , authority structures and power struggles between difference stakeholders in a postcolonial world are inevitable components , with major repercussions for the planning and implementation process . Interventions are also designed through policy narratives that define problems and solutions in specific ways; but these storylines can be overly simplistic to reduce uncertainties , appeal to ideals of feasibility and to enroll support ( see the work of Roe [47] for an overview of this position ) . For example in Zambia , the larger policy environment of the Millennial Development Goals ( MDGs ) , local government decentralization reforms , the failure of past latrine subsidy approaches and the need to “reinvent” the sanitation sector as well as a global discourse about the appeal of the CLTS technique itself ( its simplicity , low cost and impact ) motivated the rationale for its implementation . In Uganda , the use of private veterinarians to sustain sleeping sickness parasite reductions , after mass cattle treatments , was generated through an emergency narrative of the eminent merger of the two sleeping sickness forms ( Rhodesian and Gambian Human African Trypanosomiasis ( HAT ) , which is unique to Uganda ) , the synergies between business and public health and the low cost and simplicity of insecticide application . These cogent narratives were framed as locally appropriate but ended up locking themselves into certain delivery pathways involving a specific set of actors , that was difficult to modify–a pattern discussed at length by Leach et al . [48] in relation to ongoing problems with international development planning and implementation more generally . Hence the narrative helped enroll certain actors and perspectives while excluding or marginalizing others . Policy processes can drive the mobilization of resources and the arranging of intervention strategies in very linear and technocratic ways , where confidence in the intervention’s social engineering ( as the political scientist James Scott in his well-known work , Seeing Like a State [49] , would say ) becomes over-extended . In my case studies , the interventions were funded , managed and driven by different stakeholders–international agencies , academic institutions , the private sector , philanthropic foundations and a variety of government ministries ( this institutional ecosystem for NTD control is described in the review paper by [50] ) . Multiple bottlenecks in the planning , managing and governance of the interventions limited field-orientated pragmatism , most often influenced by various differences between the stakeholders involved and their ability to learn from operational mistakes and maneuver within their spheres of influence . In Zambia , the focus of UNICEF on strengthening local government decentralization led to CLTS funds and management being channeled through the Ministry of Local Government and Housing ( MLGH ) . However the MLGH rural water and sanitation department had little experience with participatory methods or in rural sanitation . In my study district , the department was staffed by recent graduates from other urban areas of Zambia who had somewhat patronizing views of “dirty villagers” , paid no attention to the past history of sanitation interventions in the district and did not have a strong desire to involve chiefs , EHTs and local volunteers , as emphasized in “true” CLTS field guides . Mismanagement , then , was a key to the low effectiveness of the project , but so were the important institutional histories , norms and conflicts between the different stakeholders . Had the district focal person for CLTS been more committed to success , and been more supported in this regard , outcomes would likely have been very different , as occurred in neighboring districts . These issues were also encountered in Tanzania and Uganda . The rabies elimination project , centrally organized by the WHO office in Dar es Salaam , distributed equally set budgets to all 28 districts irrespective of geography , infrastructure and dog populations . Kilombero and Ulanga districts , however , were some of the largest districts and a more flexible budget planning approach would have helped address many problems . The fact that the program budget was often sent at unpredictable times of the year , and needed to be used before the end of the fiscal year , obliged the district teams to use the funds when extensive flooding and pastoralist migrations had occurred , as mentioned above . Furthermore , the history of Structural Adjustment Policies ( SAPs ) ( macro-economic reforms instituted by the International Monetary Fund and World Bank in the 1980s as part of a neoliberal agenda ) on the veterinary sector in Tanzania meant that staff capacity was sub-optimal , and contributed to negative community perceptions and relationships with local vet extension officers . The top-down methods of planning used by the WHO country office maintained these rigidities . In Uganda , the interests of the private sector partners meant that the project could not easily adapt to promote a cheaper insecticide . The eventual rolling back of financial support for community education ( airtime , money for village leaders , motorcycle repairs and salaries ) was driven by the aim of creating self-sustaining businesses . Here was a narrative that sustainable veterinary business could drive sleeping sickness control . But this did not take into account the particularities of disease epidemiology , or the fact that the original spray-team model needed to be adapted . Important changes in socio-demographics and delivery during the course of the SOS interventions were also unaccounted for: a total of 80% of veterinary shops in my four study districts ( survey in 2012 ) had been established since the original business model intervention in 2008 . However the changing veterinary drug market , the continued movements of infected cattle into the area and the high amount of non-tsetse insecticides being sold were left unaddressed . These examples show that policy pathways are hard to change once they are set into motion . A lack of finances , capacity and reflexive management–itself influenced by the conceptual frameworks for action and organizational limitations–all serve to maintain existing courses of action , despite the need for adaptation and fine-tuning on the ground . There were alternative governance arrangements that could have avoided some of the institutional barriers involved—for example , rabies vaccination managed by NGOs in the Serengeti ( Tanzania ) were known as more adaptive to local circumstances; the medical sector and International NGOs were both considered more competent to implement CLTS; and not having been tied exclusively to the SOS brand insecticide through private sector partners , or working more closely with the local HAT treatment centers , could have opened up the possibility of promoting cheaper pyrethroid products , and/or tailoring approaches to sleeping sickness endemic villages . Management inertia was maintained by a lack of accountability , poor monitoring and evaluation systems and an obscuring of the true impact of the program on the ground . This is something that critical medical anthropologists , working in diverse locations around the world , have begun to unpack as they shine their ethnographic gaze on to the bureaucracy and performance of global health itself [51–52] . Intervention ownership was an important factor as well; in all cases , the interventions appeared ( and were often spoken about as being ) “imposed” by external stakeholders and were time-limited . The short project cycle appears to generate its own psychological burden that negatively influences the drive for success for local managers and field staff: why rock the boat if the project is going to end soon anyways , and it will not necessarily further your career interests ? Building up delivery networks and service provision is itself a process that , in many ways , defies the otherwise short-term goals and targets . There are also long-term challenges with capacity , state-citizen relationships , employment and governance that need to also be acknowledged .
The way we think about the world influences the way we approach social problems , like disease and poverty . The word “intervention” itself comes from the Latin , meaning a “coming-between . ” In global health and NTD control , intervening is a social and political act , one that extends biomedicine , public health and development ( physically and socio-culturally ) from the center ( where wealth , power and material advancement are greater ) to the periphery , where “diseases of poverty” , by their very definition , are predominately found and clustered [53] . The different world of the international boardroom , the district office and the village ( or pastoralist field and urban shantytown ) offer very different subjectivities . Inefficiencies , poor decisions , blind-spots , conflicts and unspoken social rules mediate the relationships between these social groups . This is a complicated process , especially as the world continues to evolve from colonial pasts into uncertain multi-polar geopolitical futures where interlocking socio-ecological global crises will present unpredictable challenges for NTD control [29–30] . There is a lot to unpack here , as the discussion above has shown , and as a synopsis of the key findings from the three case studies and their relevance to planning and M&E , show in Table 2 . To be clear , budget constraints are an important part of ineffective interventions . For example , there would have likely been a different outcome had SOS continuously subsidized insecticide distribution and funded the private veterinarians for continued social mobilization in endemic villages; the budget for rabies vaccination ensured greater capacity to vaccinate in remote areas; and CLTS paid their volunteers and enrolled the support of other stakeholders through direct financial incentives . Money alone could not have solved all of the barriers to change discussed above , but it would have certainly helped . In an ideal world , program planners would get ample operational budgets . The reality is otherwise . Budgets are also stretched from the start , because those who administer and write the grants know that making big promises ( for example , to cover large terrains ) , sometimes beyond what is really tangible , is what can secure the money from international and domestic political and scientific constituents . They also , of course , want to extend their ambitions for positive population health over as large an area as possible . These realities aside , there are also lots of ways to better use existing resources . The art of project management is to ensure success in the face of limits–a balancing act . As I have argued , one major shortcoming of current intervention approaches , whether more top-down ( the WHO rabies vaccination program ) , participatory ( CLTS in Zambia ) or public-private partnerships ( SOS in Uganda ) –is the lack of a critical praxis ( dialectical movements between reflection and action ) embedded within project planning . M&E systems are often weak . Funds are stretched , and/or used in sub-optimal ways . Existing ways of doing things go unchallenged . Hierarchies govern over best practice . Inertia and fatigue sets in . Time is short . Planners spend more time looking upwards to donors than downwards to recipients . The challenge of establishing delivery systems means that community engagement is sidelined . Learning about the details gets pushed to the side . Proactively learning about the process of implementation , using a framework that is attuned to the forensic details of the process and context itself , may be one hopeful antidote to the placidity of this daily-grind . Seeking out information on the terrain of implementation , the agency of local communities , the strategies and incentives of field staff , the socio-materiality of technology and the challenges of governance should , so I hope , assist in promoting interventions as a process of , what Biehl and Petryna [54] have ingeniously called , the task of “endlessly tinkering” intervention techniques and strategies on the ground . An intervention is never a completed process . It is always evolving and changing . It demands revision . It demands an ethical and moral engagement . Such an orientation should take place at multiple levels , including with local managers and field staff , and with the methodologies and tools of implementation itself . This should include socio-anthropologists communicating more effectively about what they can bring to the table , and be assisted in doing so by promoting a culture of problem solving , of constructive criticism , critique and adaptation [55] . In a recent book , Reimagining Global Health , Paul Farmer and colleagues [56] argued that global health needs to take a biosocial approach committed to equity and social justice . Many would agree; but the problem is , as the authors note from experience , that “no one sets out to ignore equity…the way we frame issues of causality and response typically fails to give it due consideration” [56] . This gets to the heart of how important our perspectives are in shaping our actions . One small way forward , I think , is to think more critically about what interventions are , what implementation entails and the nature of social action across diverse contexts . To this end , social science , in its ideal form , can offer us what Flyvbjerg [57] has called a “practical wisdom . ” Such an approach has much to offer the science of global health delivery , including the control of neglected tropical diseases . | Many efficacious tools exist to control NTDs , but effectively moving these tools and approaches from the boardroom to the village is a complicated socio-political process . In the era of Sustainable Development Goals , global health has become more focused on improving the delivery of existing interventions . Greater attention to implementation research , including the value of social science perspectives , has followed in an effort to build a science of global health delivery . This paper presents an accessible and actionable socio-anthropological framework for understanding the effectiveness factors of NTD interventions . The framework was developed by comparatively analyzing three large-scale NTD interventions in Eastern Africa: rabies elimination in Tanzania , sleeping sickness control in Uganda and the prevention of parasitic worms in Zambia . The framework includes five “intervention domains” where the effectiveness of these interventions was determined: 1 ) the terrain of intervention; 2 ) community agency; 3 ) the strategies and incentives of field staff; 4 ) the socio-materiality of technology; and 5 ) the governance of interventions . The paper illustrates the importance of each of these domains , presenting lessons learnt and practical recommendations . As a flexible analytical tool , the framework could be integrated into the planning and implementation process itself , bringing the insights of socio-anthropological approaches into an emerging science of NTD delivery . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"geographical",
"locations",
"vertebrates",
"uganda",
"dogs",
"animals",
"mammals",
"rabies",
"tanzania",
"global",
"health",
"neglected",
"tropical",
"diseases",
"africa",
"veterinary",
"science",
"public... | 2018 | Towards a science of global health delivery: A socio-anthropological framework to improve the effectiveness of neglected tropical disease interventions |
Effective triage of dengue patients early in the disease course for in- or out-patient management would be useful for optimal healthcare resource utilization while minimizing poor clinical outcome due to delayed intervention . Yet , early prognosis of severe dengue is hampered by the heterogeneity in clinical presentation and routine hematological and biochemical measurements in dengue patients that collectively correlates poorly with eventual clinical outcome . Herein , untargeted liquid-chromatography mass spectrometry metabolomics of serum from patients with dengue fever ( DF ) and dengue hemorrhagic fever ( DHF ) in the febrile phase ( <96 h ) was used to globally probe the serum metabolome to uncover early prognostic biomarkers of DHF . We identified 20 metabolites that are differentially enriched ( p<0 . 05 , fold change >1 . 5 ) in the serum , among which are two products of tryptophan metabolism–serotonin and kynurenine . Serotonin , involved in platelet aggregation and activation decreased significantly , whereas kynurenine , an immunomodulator , increased significantly in patients with DHF , consistent with thrombocytopenia and immunopathology in severe dengue . To sensitively and accurately evaluate serotonin levels as prognostic biomarkers , we implemented stable-isotope dilution mass spectrometry and used convalescence samples as their own controls . DHF serotonin was significantly 1 . 98 fold lower in febrile compared to convalescence phase , and significantly 1 . 76 fold lower compared to DF in the febrile phase of illness . Thus , serotonin alone provided good prognostic utility ( Area Under Curve , AUC of serotonin = 0 . 8 ) . Additionally , immune mediators associated with DHF may further increase the predictive ability than just serotonin alone . Nine cytokines , including IFN-γ , IL-1β , IL-4 , IL-8 , G-CSF , MIP-1β , FGF basic , TNFα and RANTES were significantly different between DF and DHF , among which IFN-γ ranked top by multivariate statistics . Combining serotonin and IFN-γ improved the prognosis performance ( AUC = 0 . 92 , sensitivity = 77 . 8% , specificity = 95 . 8% ) , suggesting this duplex panel as accurate metrics for the early prognosis of DHF .
Dengue is arguably the most important arboviral disease globally , with an estimated 390 million infections occurring yearly , of which nearly 100 million are clinically apparent [1] . The clinical manifestations of dengue infections range from mild undifferentiated febrile illness to classical dengue fever ( DF ) , severe dengue hemorrhagic fever ( DHF ) and sometimes fatal disease . The two predominant pathophysiologic observations in DHF are increased vascular permeability and thrombocytopenia , which result in plasma leakage and increased risk of hemorrhage , respectively . Early triage in the acute , febrile phase is useful for streamlining case management and monitoring . One of the pathophysiological hallmarks of severe dengue is increased vascular permeability , leading to a loss in blood volume that if uncorrected , could lead to shock . The degree of plasma leakage varies and prediction of severe plasma leakage has been challenging [2] . Daily or even more frequent monitoring of hematocrit levels may be required for at least 2–3 days around the period of fever defervescence to detect plasma leakage soon after its onset . As timely fluid support prevents progression of plasma leakage to hypovolemic shock and multi-organ failure , tools that enable triaging of patients according to disease outcome could enable a more effective use of limited healthcare resources , especially during dengue epidemics [3] . Indeed , the components of early pathophysiological mechanisms could serve as reliable prognostic factors given their mechanistic role in disease manifestation . It is generally believed that the observed pathophysiology is immune-mediated brought about by monocytes , T-cells , endothelial cells , mast cells and increasingly platelets , as well as the interactions between these cell types [4–7] . Despite the consistent presence of thrombocytopenia in the acute stages of dengue , little research has focused on the pathologic contribution of platelet-derived soluble factors in DHF progression and their use as prognostic markers . Platelets circulate in high numbers throughout the system and upon activation release their granule contents to exert their hemostatic , immunological and inflammatory effects [8 , 9] . DENV has been reported to activate platelets [6] . One of these bioactive compounds released during platelet activation is serotonin ( 5-hydroxytryptamine or 5-HT ) . Platelets store serotonin in the dense granules but lack the enzymes to synthesize serotonin . Instead , platelets take up plasma serotonin produced by enterochromaffin cells in the gastrointestinal tract , through the serotonin receptor ( SERT or 5-HTT ) [10] . Serotonin is secreted during platelet activation and promotes platelet aggregation . It also further amplifies platelet activation and aggregation through 5-HT2A receptor re-uptake to result in vasoconstriction of surrounding blood vessels and hemostasis [11] . While in vitro results suggest DENV induced platelet apoptosis [6] , the reduction of platelet number and function has not been characterized in vivo . Furthermore , thrombocytopenia and associated platelet dysfunction is short-lived and the rapid recovery of platelet numbers in the convalescence phase suggest that the function of soluble metabolites and/or immune mediators with short half lives may play critical roles in disease progression . In this study , we adopted a metabolomics to study dengue-induced metabolites , especially platelet-derived molecules and evaluated their potential as prognostic biomarkers for severe dengue . We discovered a major reduction in circulating serotonin in both DF and DHF , with the reduction in DHF exceeding that of DF . In addition , serotonin levels correlated with the degree of thrombocytopenia , and when used in combination with IFN-γ , they provide accurate early ( <96 h from onset of fever ) prognosis of DHF .
The dengue study cohort of 116 dengue patients– 60 DF patients and 56 DHF patients ( Table 1 ) were recruited from the Prospective Adult Dengue Study ( PADS ) [12] . Briefly , PADS is a cohort study of acutely febrile adults at a tertiary care center , Communicable Diseases Center , Tan Tock Seng Hospital , Singapore . Adult patients ( ≥ 18 years ) presenting with acute onset of fever ( ≥ 37 . 5°C ) without rhinitis or other clinical alternatives were included in the study ( Febrile stage , < 96 hours post onset of fever; Defervescence , Day 5–7 , Convalescence , Day 21–28 ) . Venous blood samples were collected , aliquoted and frozen at -80°C for hematological , virological and serological analysis . Enrollment of all eligible individuals was based on written informed consent and the collected samples were anonymized . The protocols were approved by the Domain Specific Review Board of the National Healthcare Group , Singapore ( DSRB/E/2009/432 ) . We reported our study design , hypotheses , patient characteristics , assay methods , statistical methods and modeling methods as per REMARK which is important for generalizability [13] . Additionally we used serum samples from 24 asymptomatic age- and gender-matched healthy subjects as controls . DF and DHF patients were classified according to the WHO 1997 dengue guidelines [14] . To fulfill the case definition of DHF , all four of the following criteria must be present , namely: fever or history of fever , hemorrhagic tendencies , thrombocytopenia and evidence of plasma leakage [14] . Hematoconcentration was determined by the hematological analyzer and expressed as % of the volume of whole blood that was made up of red blood cells . Hematocrit increase of over 20% of the values at convalescence phase is considered a common clinical index of plasma leakage and DHF diagnosis . The PADS cohort comprised of both DF and DHF patients recruited at different phases of dengue infection– ( febrile phase: DF = 25; DHF = 27; defervescence phase: DF = 31; DHF = 29 , convalescence phase: DF = 25; DHF = 25 ) . DENV2 is the predominant DENV type . Based on the anti-DENV IgG and IgM seropositivity or seronegativitiy , immune status of the patients was determined . Anti-DENV IgM seropositivity and IgG seronegativity indicated that 40–52% of DF and 22–41% of DHF are primary cases , and the remaining secondary cases . 12–16% DF patients and 59–76% DHF patients had platelets <50×103/μL at any point as determined during their daily routine total blood count . A detailed hematological and virological analysis was performed and included white blood cell count ( WBC ) , red blood cell count ( RBC ) , blood hemoglobin ( HGB ) , hematocrit ( HCT ) , mean corpuscular volume ( MCV ) , mean corpuscular hemoglobin ( MCH ) , mean corpuscular hemoglobin concentration ( MCHC ) , platelet count ( PLT ) , lymphocyte percentage ( LYMPH% ) , lymphocyte count ( LYMPH ) , mixed cell count ( MXD ) , neutrophil percentage ( NEUT% ) , neutrophil count ( NEUT ) , red blood cell distribution width-coefficient of variation ( RDW-CV ) , and quantitation of peripheral viral titers using reverse transcriptase-polymerase chain reaction ( RT-PCR ) crossover values ( Ct ) . Dengue viral infection was confirmed by RT-PCR [15] , or NS1 detection by Dengue NS1 Ag Strip ( Bio-Rad , Marnes-la-Coquette , France ) at the Environmental Health Institute , Singapore , or typing by virus isolation and immunofluorescence using DENV type-specific monoclonal antibodies ( ATCC: HB46-49 ) . Dengue-immune status ( primary or secondary DENV infection ) was based on Dengue IgG levels in the acute sera , using a commercially obtained ELISA ( PanBio , Brisbane , Australia ) according to the manufacturer’s protocol . For untargeted metabolomics analysis , a volume of 50 μL from each serum sample was thawed at 4°C and serum proteins were precipitated with 200 mL ice-cold methanol , which contained 10 mg/mL 9-fluorenylmethoxycarbonyl-glycine as an internal standard . After vortexing , the mixture was centrifuged at 16 , 000 rpm for 10 minutes at 4°C and the supernatant was collected and evaporated to dryness in a vacuum evaporator . The dry extracts were then redissolved in 200 μL of 98:2 water/methanol for liquid chromatography-mass spectrometry ( LC-MS ) analysis . Quality control ( QC ) samples were prepared by mixing equal amounts of serum samples from all the samples and processed as per other samples . The QC sample was run after each 8 samples to monitor the stability of the system and all samples were randomized . For targeted metabolomics analysis , sample preparation followed a published report with some modifications [16] . Briefly , 10 μL of the internal standard mix was added to 50 μL of serum . The sample was then diluted to 100 μL with water containing 0 . 1% formic acid ( v/v ) , vortexed , followed by the addition of 400 μL of cold methanol . After vortexing , the mixture was centrifuged at 16 , 000 rpm for 10 minutes at 4°C and the supernatant was collected and evaporated to dryness in a vacuum evaporator . The dry extracts were then redissolved in 200 μL of 0 . 1% formic acid in water for LC-MS/MS analysis . A summary of the workflow utilized in untargeted and targeted metabolomics studies is shown in S1 Fig . Untargeted metabolomics were performed as previously described with modifications [17] . The supernatant fraction from sample preparation step was analyzed using Agilent 1290 ultrahigh pressure liquid chromatography system ( Waldbronn , Germany ) equipped with a 6520 QTOF mass detector managed by a MassHunter workstation . The column used for the separation was an Agilent rapid resolution HT Zorbax SB-C18 ( 2 . 1×100 mm , 1 . 8 mm; Agilent Technologies , Santa Clara , CA , USA ) . The oven temperature was set at 45°C . The gradient elution involved a mobile phase consisting of ( A ) 0 . 1% formic acid in water and ( B ) 0 . 1% formic acid in methanol . The initial condition was set at 5% B . A 7 min linear gradient to 70% B was applied , followed by a 12 min gradient to 100% B which was held for 3 min , then returned to starting conditions over 0 . 1 min . Flow rate was set at 0 . 4 ml/min , and 5 mL of samples was injected . The electrospray ionization mass spectra were acquired in both positive and negative ion mode . Mass data were collected between m/z 100 and 1000 at a rate of two scans per second . The ion spray voltage was set at 4 , 000 V , and the heated capillary temperature was maintained at 350°C . The drying gas and nebulizer nitrogen gas flow rates were 12 . 0 L/min and 50 psi , respectively . Two reference masses were continuously infused to the system to allow constant mass correction during the run: m/z 121 . 0509 ( C5H4N4 ) and m/z 922 . 0098 ( C18H18O6N3P3F24 ) . The targeted LC-MS/MS analysis followed a published report with some modifications [16] . Briefly , LC-MS analysis was performed with Agilent 1290 ultrahigh pressure liquid chromatography system ( Waldbronn , Germany ) coupled to an electrospray ionization with iFunnel Technology on a triple quadrupole mass spectrometer . Chromatographic separation was achieved by using Atlantis T3 column ( 2 . 1×100 mm , 3 μm; Waters , Eschbornn , Germany ) with mobile phases ( A ) 0 . 1% formic acid in water and ( B ) 0 . 1% formic acid in methanol . The initial condition was set at 0% B . A 10 min linear gradient to 40% B was applied , followed by 1 min gradient to 100% B which was held for 5 min , then returned to starting conditions over 0 . 1 min . The column was kept at 40°C and the flow rate was 0 . 4 mL/min . The auto-sampler was cooled at 4°C and an injection volume of 5 μL was used . Electrospray ionization was performed in positive ion mode with the following source parameters: drying gas temperature 200°C with a flow of 14 L/min , nebulizer gas pressure 30 psi , sheath gas temperature 400°C with a flow of 11 L/min , capillary voltage 3 , 000 V and nozzle voltage 800 V . Compounds were quantified in multiple reaction monitoring ( MRM ) mode with the following transitions: m/z 177>160 , m/z 181>164 , m/z 209>192 , m/z 213>196 , m/z 233>174 , m/z 192>146 , and m/z 213>196 for serotonin , d4-serotonin , kynurenine , and d4-kynurenine , melatonin , 5-hydroxy-indole-3-acetic acid ( HIAA ) , and 5-hydroxytryptophan ( HTP ) , respectively . Data acquisition and processing were performed using MassHunter software ( Agilent Technologies , US ) . A representative LC-MS/MS chromatogram of a native standards mix is shown in S2 Fig . The method was validated for limit of detection ( LOD ) , linearity , accuracy , precision and recovery , according to Food and Drug Administration ( FDA ) guidelines for biological method as previous published report [16] . Briefly , the calibration curves were constructed from three replicate measurements of eight concentrations of each standard . A linear regression with r2 >0 . 995 was obtained in all relevant ranges . The LODs , defined by a signal-to-noise ratio ( S/N ) of 3 , ranged from 0 . 5 to 10 . 0 nM for all the analytes . The recoveries were evaluated by spiking defined amounts of analytes into aliquots of unprocessed serum and analyte concentrations were calculated using the calibration curves . The spiked concentration was obtained by subtracting the endogenous concentration which was determined from the analysis of the unspiked sample . The recoveries generally ranged from 52 . 4% to 82 . 6% . For intra-batch and inter-batch precision and accuracy , the relative standard deviation ( RSD ) values ranged from 1 . 1% to 13 . 2% all the analytes . Fluorescent bead based measurement of cytokines , chemokines and growth factors in the patients’ sera were performed in duplicates using the Luminex technology xMAP ( Bioplex 27-plex human cytokine kit , Bio-Rad , California , USA ) as per manufacturer’s instructions . The measured analytes are IL-1β , IL-1ra , IL-2 , IL-4 , IL-5 , IL-6 , IL-7 , IL-8 , IL-9 , IL-10 , IL-12 , IL-13 , IL-15 , IL-17 , basic FGF , Eotaxin , G-CSF , GM-CSF , IFN-γ , IP-10 , MCP-1 ( MCAF ) , MIP-1α , MIP-1β , PDGF-BB , RANTES , TNF-α and VEGF . The standard curves were optimized automatically by the software ( Bioplex manager ) and verified manually . In order to prevent batch effect , samples were randomized prior to analysis . Calibrations and validations were performed prior to analyses . Raw spectrometric data in untargeted metabolomics were analyzed by MassHunter Qualitative Analysis software ( Agilent Technologies , US ) and the molecular features characterized by retention time ( RT ) , chromatographic peak intensity and accurate mass , were obtained by using the Molecular Feature Extractor algorithm . The features were then analyzed by MassHunter Mass Profiler Professional software ( Agilent Technologies , US ) . Only features with an intensity ≥ 20 , 000 counts ( approximately three times the limit of detection of our LC-MS instrument ) , and found in at least 80% of the samples at the same sampling time point signal were kept for further processing . Next , a tolerance window of 0 . 15 min and 2 mDa was used for alignment of RT and m/z values , and the data normalized to spiked 9-fluorenylmethoxycarbonyl-glycine internal standard . Raw spectrometric data in targeted metabolomics were processed using MassHunter Workstation Quantitative Analysis software ( Agilent Technologies , US ) . For statistical analysis , nonparametric Test ( Wilcoxon , Mann–Whitney test ) with Benjamini-Hochberg Multiple Testing Correction was employed , because the samples analyzed were obtained from different patients , and statistical significance was set at p<0 . 05 . For multivariate data analysis using hierarchical clustering or Orthogonal projections to latent structures discriminant analysis ( OPLS-DA ) , data were normalized by median-centering and dividing by standard deviation . Hierarchical clustering was performed using MeV version 4 . 9 . 0 . OPLS-DA was performed using the software package SIMCA-P 13 . 0 version ( Umetrics AB , Umea , Sweden ) . Metabolites and cytokines/chemokines with Variable Importance in the Projection ( VIP ) values>1 were considered to be influential for the separation of samples in OPLS-DA analysis . In addition , the fold change ( FC ) analysis was also performed to further filter the features and only those features with FC > 1 . 5 were selected as potential significantly altered metabolites . Receiver operating characteristic ( ROC ) curve was made by using R package . The structure identification of the differential metabolites was based on our published work [17] . Briefly , the elemental compositions of the metabolites were first calculated based on the exact mass , the nitrogen rule and the isotope pattern by Masshunter software from Agilent . Then , the elemental composition and exact mass were used for open source database searching , including LIPIDMAPS ( http://www . lipidmaps . org/ ) , HMDB ( http://www . hmdb . ca/ ) , METLIN ( http://metlin . scripps . edu/ ) and MassBank ( http://www . massbank . jp/ ) . Next , MS/MS experiments were performed to obtain structural information via the interpretation of the fragmentation pattern of the metabolite . The MS/MS spectra of possible metabolite candidates in the databases were also searched and compared . Finally , the metabolites were confirmed by comparison with the standards where commercially available , which was the case for serotonin and kynurenine . The metabolites are listed according to the minimum reporting standards for chemical analysis in metabolomics recommended by Metabolomics Standard Initiative ( MSI ) [18 , 19] . Briefly , a four-level system ranging from Level 1 ( identified metabolites ) via Levels 2 and 3 ( putatively annotated compounds and compound classes ) to Level 4 ( unidentified or unclassified metabolites which can still be differentiated based on spectrum data ) .
We characterized the metabolome changes early in dengue infections using liquid-chromatography tandem mass spectrometry ( LC-MS/MS ) to globally map the serum metabolomes from DF ( n = 25 ) and DHF ( n = 27 ) patients . In this untargeted , global metabolomics , a total of 20 MSI Levels 1 and 2 metabolites were significantly different between DHF and DF patients in the febrile phase , of which 8 were increased [ ( L-kynurenine , 13E-docosenamide , deoxyinosine , N-Heptanoylglycine , 3-carboxy-4-methyl-5-propyl-2-furanpropionic acid , bilirubin , phosphotidylethanolamine ( 20:4/20:3 ) , phosphatidylserine ( 18:0/18:0 ) and 12 were decreased ( Leucyl-phenylalanine , phenylalanyl-tryptophan , leucyl-alanine , palmitic amide , serotonin , oleamide , phenylalanylphenylalanine , lysophosphotidylethanolamine ( 20:0/0:0 ) , lysophosphotidylethanolamine ( 22:4/0:0 ) , phosphatidylserine ( 18:0/20:0 ) , phosphotidylethanolamine ( 20:4/P-18:1 ) , phosphotidylethanolamine ( 18:1/22:5 ) ] in DHF patients relative to DF patients ( Table 2 ) . MSI Levels 3 and 4 metabolites were listed in S1 Table . Hierarchical clustering based on the metabolome profile optimally segregated the patients ( Fig 1 ) , where all but four DF patients and three DHF patients were classified separately , therefore correlating well with WHO 1997 dengue classification scheme . Verification of peaks’ retention time and MS/MS spectrum through the use of synthetic standards provided highly confident identification of serotonin and kynurenine . Among the hematological parameters that are routinely tested in the clinics , platelets correlated most frequently with the metabolites , in particular , demonstrating significant positive correlations to serotonin ( r = 0 . 67; p<0 . 0001 ) and negative correlations to kynurenine ( r = ‒0 . 45; p<0 . 005 ) ( Fig 2 ) . Decrease in serotonin levels was accompanied with an increase in kynurenine levels in the febrile phase ( Fig 3A and 3B ) . An increase in indoleamine 2 , 3-dioxygenase in DF patients was reported previously [20] , and is presumably the source of kynurenine . We further evaluated serotonin and kynurenine levels in order to better understand their levels across the course of dengue infection . To improve analytical specificity for absolute quantitative determinations , we developed a high-throughput precision assay based on stable-isotope dilution mass spectrometry of serotonin and kynurenine . Using defined concentrations of deuterated internal standards spiked into samples and calculating the response ratio of the analyte of interest to the internal standard , we accurately determined the concentrations of serotonin and kynurenine in DF and DHF patients ( Fig 4A and 4B ) . Table 3 summarizes the mean values of serotonin and kynurenine ( nM ) at the three phases in DF and DHF patients . Compared to DF patients , we observed significant decreases in serotonin in DHF patients in both the febrile ( p<0 . 001 ) and the defervescence phases ( p<0 . 001 ) , but not the convalescence phase . Kynurenine on the other hand was significantly different between DF and DHF patients in the febrile phase ( p<0 . 01 ) , but not in the defervescence and convalescence phases . Furthermore , a dynamic change of serotonin levels was observed in both DF and DHF patients in all three phases of infection . Compared to febrile stage , the level of serotonin continued to decrease and reached the lowest level at defervescence phase . At convalescence phase , increased serotonin was observed and the level was significantly higher than the levels in both febrile and defervescence phases . Meanwhile , there was no difference in serotonin levels between cases of primary and secondary infection in all three phases ( S3 Fig ) . Similar to the correlation results in global-scale metabolomics ( Fig 2 ) , significant positive correlation between serotonin concentrations and platelet numbers ( r = 0 . 55 , p = 0 . 0070 ( DF ) and r = 0 . 67 , p = 0 . 0001 ( DHF ) ; S4A Fig ) was observed in the febrile phase , suggesting that the decrease in serotonin is associated with decreased platelet numbers in DF and DHF patients . Interestingly , the correlation between serotonin and platelets ceased in the defervescence phase ( r = 0 . 20 , p>0 . 05; S4B Fig ) . If serotonin levels and platelet counts were organized according to different fever days of dengue patients , serotonin levels continued to decrease from onset of fever up to Day 6 and 7 ( S4C Fig ) , which did not parallel the initial decrease then recovery of platelet numbers with time ( S4D Fig ) . The levels of HIAA and melatonin , two main degradation products of serotonin , and HTP , the precursor of serotonin , were also evaluated , by using d4-serotonin as their internal standard . While the levels of melatonin in serum were below the detection limit , the concentrations of HTP and HIAA in DF and DHF patients were determined ( S5A and S5B Fig ) . Unlike serotonin , neither HTP nor HIAA showed any significant difference between DF and DHF in both the febrile and the defervescence phases . Given the early significant changes of serotonin and kynurenine in DHF patients , and the numerous roles of serotonin in platelet aggregation and activation [11] and kynurenine in immunomodulation [21] , Receiver Operating Curve analyses were performed to assess their prognostic utility . The higher performing AUC of serotonin ( AUC = 0 . 80 , 95% C . I . 0 . 68–0 . 92 , p = 0 . 0002 ) shows its prognostic superiority compared to kynurenine ( AUC = 0 . 72 , 95% C . I . 0 . 57–0 . 86 , p = 0 . 008 ) ( S6A and S6B Fig ) , and was henceforth selected as a better prognostic biomarker for DHF . To improve on serotonin predictive ability and extend the scope of capturing inflammatory compounds in dengue infections as prognostic biomarkers of DHF , we performed multiplex immunoassays on 27 cytokines/chemokines . Nine were significantly different between DF and DHF namely , IFN-γ , IL-1β , IL-4 , IL-8 , G-CSF , MIP-1β , FGF basic , TNFα and RANTES ( S7 Fig ) . VIP plots generated from OPLS-DA modelling revealed IFN-γ as the top ranked cytokine in dengue infections ( S8A Fig ) . The importance of IFN-γ as a prognostic biomarker is consistent with it being widely reported as an important pro-inflammatory cytokine in dengue [22 , 23] . In addition , due to reports of the elevation of IL-10 in DHF patients , IL-10 has been suggested as a biomarker of severe dengue [24 , 25] . We compared the DHF prediction potential for IFN-γ alone , IL-10 alone , platelets alone , and the combination of serotonin and IFN-γ at <96 h from onset of fever . The AUCs of platelets , serotonin alone , IFN-γ alone and IL-10 alone were 0 . 78 ( 95% C . I . 0 . 66–0 . 90 , p = 0 . 0001 ) , 0 . 80 ( 95% C . I . 0 . 68–0 . 92 , p = 0 . 0002 ) , 0 . 88 ( 95% C . I . 0 . 79–0 . 97 , p<0 . 0001 ) , and 0 . 55 ( 95% C . I . 0 . 40–0 . 69 ) , respectively ( S8B and S8C Fig ) . We combined serotonin and IFN-γ which resulted in AUC of 0 . 92 , sensitivity = 77 . 8% and specificity = 95 . 8% ( p<0 . 0001; Fig 5 ) in predicting DHF .
Through a combined untargeted and targeted metabolomics screen in dengue patients , we have identified depressed levels of circulating serotonin in DF and DHF patients . Lower serotonin levels in DF patients have been observed previously and are consistent with our study [26] . Our study thus extended on previous knowledge by demonstrating that the declines of serum serotonin are steeper in DHF patients compared to DF patients in the febrile phase and this declining trend continues into the defervescence phase . Due to early drop in circulating serotonin within the first 96 h from onset of fever , we propose that serotonin , and the inclusion of cytokines , such as IFN-γ may be used as prognostic biomarkers for the early prognosis of DHF . Circulating serotonin forms a distinct pool separate from central nervous system serotonin pool , and is taken up via SERT and stored in platelets . During platelet aggregation , platelet-stored serotonin is released into circulation [27] , which in turn promotes platelet aggregation in a feedback fashion via the serotonin receptor ( 5-HT2A ) on platelets [28] . It has been shown that sera from DHF patients cross-react with platelets and inhibit platelet aggregation due to the auto-antibodies directed against DENV nonstructural protein 1 ( NS1 ) [29 , 30] . Anti-NS1 antibodies bind to platelet membrane protein disulfide isomerase ( PDI ) [31] . Moreover , these anti-NS1 autoantibodies induce platelet lysis [30] . Given that systemic NS1 protein levels correlate with viremia and dengue severity [32 , 33] , it is conceivable that attenuated platelet aggregation and numbers in dengue patients may have resulted in the depressed serum serotonin levels and even more so in DHF patients . In addition , animal models and human ex vivo studies have shown that NS1 cross-reacts with platelets and endothelial cells and reduce platelet aggregation through the generation of auto-anti-platelet antibodies [30 , 31 , 34] . Notably , similar levels of HTP , the precursor and HIAA , the main metabolite of serotonin , were found in DF and DHF , suggesting aberrations in the release and/or uptake of serotonin rather than alterations in serotonin metabolism . The targeting of SERT by serotonin-selective reuptake inhibitors ( SSRIs ) which inhibit the uptake and storage of platelet-serotonin is known to decrease platelet aggregation responses and consequently increase bleeding time [35] . Indeed , plasma leakage is the defining feature of severe dengue [14] and platelets are critical in maintaining the integrity of the vascular system . The steeper declines in serum serotonin in DHF patients in both febrile and defervescence phases suggest why plasma leakage occurs in DHF patients rather than DF patients . Therefore , our study provides mechanistic clues to how thrombocytopenia , steep serotonin decrease and plasma leakage may be linked in severe dengue . In dengue , the mechanisms leading to thrombocytopenia are poorly understood and could occur via several modes , including peripheral platelet destruction by the host immune system [6 , 36–38] , bone marrow aplasia [39] , aberrant platelet function or signaling [6] , or all of the above . Recent studies reveal that platelets participate in inflammation by influencing adaptive immunity through interactions with monocytes , neutrophils and endothelial cells [8 , 40] . In addition several studies have suggested its role in maintaining patent capillary barrier . In experimental models , platelets adhere to endothelial cells infected with DENV2 [41] , and directly interact with monocytes and neutrophils [42] . Neutrophils recruited to injured vessels under pathogenic inflammatory conditions selectively capture activated platelets [43] . In humans , platelets were reported to form platelet-leukocyte and platelet-monocyte aggregates [44] . Serotonin is cross-linked to a variety of platelet surface adhesion proteins and clotting factors [45 , 46] required for platelet aggregation and interaction with other cell types . It is not clear , however , how diminished serotonin in DF and DHF patients affects these platelet-cell interactions . Given the more rapid decline in serum serotonin in DHF patients compared to DF patients as early as 96 h after fever onset , we propose serum serotonin as a predictive marker of severe dengue prognosis . We and others previously identified IFN-γ as a candidate early prognosis biomarker [22 , 47] and we integrated serotonin and IFN-γ to achieve early prediction of patients likely to develop DHF at sensitivity of 77 . 8% and specificity of 95 . 8% within 96 h of fever onset . Our proposed serotonin with IFN-γ duplex biomarker panel attained the same prognostic performance of 0 . 92 compared to an eight feature panel previously identified [47] . This may simplify prognosis in a clinical setting . Serotonin and IFN-γ reflects the pathobiology of dengue-mediated thrombocytopenia and systemic inflammation respectively , and this information makes these candidate biomarkers biologically significant and plausible in their reflection of the pathognomonic symptoms of severe dengue . In this study , we show that serotonin levels in DHF declined more than DF patients in the febrile phase and continued to stay suppressed in the defervescence phase . Using circulating serotonin as a dengue prognosis biomarker appears to have its benefits–we demonstrated that its levels and association with thrombocytopenia in dengue is independent of whether the infection is primary or secondary . Future technological developments into rapid and cheaper analytical methods of serotonin and IFN-γ levels may facilitate early prognosis even in dengue-endemic , resource-poor areas lacking laboratory facilities , although the predictive performance of serotonin and IFN-γ needs further validation in a separate cohort . Measuring serum serotonin levels in other febrile , acute infectious disease , as well as in patients infected with different DENV type may aid in further showing specificity and potential universality of serotonin as a reliable biomarker . | Dengue , an acute arboviral disease has emerged globally , inflicting debilitating symptoms in 96 million people . The early prediction of severe dengue ( dengue hemorrhagic fever , DHF ) is challenging due to varied and late-presenting symptoms , requiring frequent monitoring for signs of disease progression . An often used parameter to monitor disease progression is decrease in platelet count , or thrombocytopenia , which is a feature of DHF . However , whether platelet-derived compounds are useful as early biomarkers predictive of DHF has not been investigated . In this study , we investigated the utility of serum metabolites as predictive biomarkers . We developed quantitative and high-throughput tools and discovered circulating serotonin , conceivably platelet-derived , that showed a nearly two-fold decrease in DHF patients compared to mild dengue fever . Because immune mediators may increase the predictive ability , we measured them in blood and identified interferon-gamma as an important cytokine in DHF . When serotonin is used in combination with IFN-γ , this dual-panel predictive panel provides accurate prognosis of DHF within 96 h from fever onset . These findings may have important clinical implications not just in early dengue prognostication but also in the design of therapeutic strategies against dengue infections . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"platelet",
"aggregation",
"tropical",
"diseases",
"biomarkers",
"neuroscience",
"serotonin",
"metabolomics",
"metabolites",
"signs... | 2016 | Serum Metabolomics Reveals Serotonin as a Predictor of Severe Dengue in the Early Phase of Dengue Fever |
Sand fly saliva has an array of pharmacological and immunomodulatory components , and immunity to saliva protects against Leishmania infection . In the present study , we have studied the immune response against Lutzomyia intermedia saliva , the main vector of Leishmania braziliensis in Brazil , and the effects of saliva pre-exposure on L . braziliensis infection employing an intradermal experimental model . BALB/c mice immunized with L . intermedia salivary gland sonicate ( SGS ) developed a saliva-specific antibody response and a cellular immune response with presence of both IFN-γ and IL-4 . The inflammatory infiltrate observed in SGS-immunized mice was comprised of numerous polymorphonuclear and few mononuclear cells . Mice challenged with live L . braziliensis in the presence of saliva were not protected although lesion development was delayed . The inoculation site and draining lymph node showed continuous parasite replication and low IFN-γ to IL-4 ratio , indicating that pre-exposure to L . intermedia saliva leads to modulation of the immune response . Furthermore , in an endemic area of cutaneous leishmaniasis , patients with active lesions displayed higher levels of anti-L . intermedia saliva antibodies when compared to individuals with a positive skin test result for Leishmania . These results show that pre-exposure to sand fly saliva plays an important role in the outcome of cutaneous leishmaniasis , in both mice and humans . They emphasize possible hurdles in the development of vaccines based on sand fly saliva and the need to identify and select the individual salivary candidates instead of using whole salivary mixture that may favor a non-protective response .
Protozoan parasites of the genus Leishmania cause a broad spectrum of diseases , collectively known as leishmaniasis , that occur predominantly in tropical and subtropical regions . The leishmaniases are transmitted by different species of sand flies , and depending on the Leishmania species involved and the genetic makeup or immunological status of the host , different clinical manifestations of the disease are observed . The mammalian host acquires leishmaniasis when it is bitten by an infected sand fly vector . During parasite inoculation , the host is also injected with the sand fly saliva which has been shown to contain a repertoire of bio-active molecules capable of interfering with the host's hemostatic , inflammatory and immune responses ( reviewed in [1] , [2] , [3] ) . Among the latter , it has been shown that sand fly saliva can exert a direct effect upon the function of cells from the immune system [4] , [5] , [6] . In fact , it was shown early on that co-inoculation of L . longipalpis or P . papatasi salivary gland sonicate ( SGS ) and L . major led to a significant exacerbation of lesion size and parasite load in BALB/c mice [7] , [8] . Similar effects were observed with L . braziliensis [9] , [10] and L . amazonensis [11] . On the other hand , mice are protected against L . major when immunized with SGS [8] , when pre-exposed to the bites of uninfected sand flies [12] or , more recently , when immunized with saliva components [13] , [14] . Since the composition of salivary molecules , their function and antigenicity varies among distinct sand fly species [15] , [16] , [17] it is of foremost importance to investigate whether vector-based vaccines can also be developed against other Leishmania species . American cutaneous leishmaniasis , caused by Leishmania braziliensis , is distinguished from other leishmaniases by its chronicity , latency and tendency to metastasize in the human host leading to muco-cutaneous leishmaniasis [18] , [19] . In this sense , we have , in the present report , investigated ( a ) the immunogenic properties of L . intermedia saliva , the main vector of L . braziliensis in Brazil , ( b ) the effect of vaccination with L . intermedia saliva followed by challenge with L . braziliensis plus saliva and c ) anti-saliva immune response of individuals from an endemic area of cutaneous leishmaniasis . In order to do so , we employed a recently developed experimental model of cutaneous leishmaniasis ( CL ) , based upon the inoculation of L . braziliensis parasites into the dermis of BALB/c mice [20] .
Female BALB/c mice ( 6–8 weeks of age ) were obtained from CPqGM/FIOCRUZ Animal Facility where they were maintained under pathogen-free conditions . All procedures involving animals were approved by the local Ethics Committee on Animal Care and Utilization ( CEUA - CPqGM/FIOCRUZ ) . Lutzomyia intermedia , Corte de Pedra strain , and Lutzomyia longipalpis , Cavunge strain , were reared at Centro de Pesquisas Gonçalo Moniz-FIOCRUZ , as described elsewhere [21] . Adult sand flies were used for dissection of salivary glands at 3–5 days after emergence . Salivary glands were stored in groups of 20 pairs in 20 µl NaCl ( 150 mM ) Hepes buffer ( 10 mM , pH 7 . 4 ) , at −70°C . Immediately before use , salivary glands were disrupted by ultrasonication within 1 . 5 ml conical tubes . Tubes were centrifuged at 10 , 000×g for 2 min and the resultant supernatant ( Salivary Gland Sonicate - SGS ) was used for the studies . The level of LPS contamination of SGS preparations was determined using a commercially available LAL Chromogenic Kit ( QCL-1000 , Lonza Bioscience ) ; LPS concentration was <0 . 1 ng/ml and SGS did not stimulate human monocytes . In order to evaluate the immunogenic potential of L . intermedia saliva , BALB/c mice were immunized three times with SGS ( equivalent to 1 pair of salivary glands ) in 10 µl of PBS , in the dermis of the right ear , using a 27 . 5 G needle . Immunizations were performed at 2 week intervals . Control mice were injected with PBS . ELISA microplates were coated overnight at 4°C with 50 µl SGS diluted to 5 pairs of salivary glands/ml in coating buffer ( NaHCO3 0 . 45 M , Na2CO3 0 . 02 M , pH 9 . 6 ) . After washing with PBS-Tween , wells were blocked with PBS-Tween plus 5% dried skim milk for 1 hour at 37°C . Wells were incubated overnight with sera from control or immunized mice , obtained two weeks after the last immunization , diluted ( 1∶50 ) in PBS-Tween . After further washings , wells were incubated with alkaline phosphatase-conjugated anti mouse IgG antibody ( Promega ) diluted ( 1∶5000 ) in PBS-Tween , for 1 hour at 37°C . Following another washing cycle , wells were developed with p-nitrophenylphosphate in sodium carbonate buffer pH9 . 6 with 1 mg/ml of MgCl2 . The absorbance was recorded at 405 nm . Serum IgG subclasses were determined using anti–mouse IgG1 , IgG2a or IgG2b alkaline-phosphatase conjugates ( Sigma ) . To check that anti-IgG1 , anti-IgG2a and anti-IgG2b were working properly , purified mouse IgG1 ( clone CD28 . 2 [2 µg/mL] ) ; mouse IgG2a ( clone HIT3a [2 µg/mL] ) and mouse IgG2b ( clone G265-5 [2 µg/mL] ) , all from BD Pharmingen ) were employed as positive controls . SDS-PAGE and Western blot of L . intermedia and L . longipalpis SGS was performed as described elsewhere [14] . Briefly , SDS-PAGE was performed by electrophoresis of SGS equivalent to 20 pairs of L . intermedia salivary glands ( ∼10 ug of protein ) and SGS equivalent to 10 pairs of L . longipalpis salivary glands ( ∼10 ug of protein ) in Bis-tris gels ( 4–12% , 1 . 0 mm ) ( Invitrogen ) . After transfer to nitrocellulose , the membrane was cut into strips , blocked overnight with blocking buffer ( Tris HCl pH 8 . 0 NaCl 150 mM plus 5% non-fat milk ) , then incubated with mouse immune serum ( 1∶50 dilution ) in blocking buffer , followed by the anti-mouse IgG alkaline phosphatase conjugate ( 1∶5000 ) ( Promega ) . Alternatively , SDS-PAGE of L . intermedia SGS ( 100 pairs of L . intermedia salivary glands or ∼50 ug of protein ) was performed on Bis-tris gels ( 4–12% , 1 . 0 mm×2D ) ( Invitrogen ) . After transfer to nitrocellulose , the membrane was blocked as above , incubated with human sera ( 1∶50 dilution ) , followed by the anti-human IgG alkaline phosphatase conjugate ( 1∶1000 ) ( Sigma ) . Bands were visualized by adding alkaline phosphatase substrate ( Promega ) . Following three intra-dermal inoculations with PBS or L . intermedia SGS in the right ear dermis , mice received a challenge injection with SGS ( equivalent to 1 pair of salivary glands ) in the left ear dermis . Twenty-four hours later , challenged ears were removed and fixed in 10% formaldehyde . Following fixation , tissues were processed , embedded in paraffin and 5 µm sections were stained with hematoxylin and eosin ( H & E ) and analyzed by light microscopy . Reagents for staining cell surface markers and intracellular cytokines were purchased from BD Biosciences , San Diego , CA . Measurement of in vitro cytokine production was performed as described in de Moura et al . [20] . Briefly , cells were activated in the presence of anti-CD3 ( 10 µg/ml ) and anti-CD28 ( 10 µg/ml ) and were later incubated with Brefeldin A ( Sigma ) ( 10 µg/ml ) . Cells were blocked with anti-Fc receptor antibody ( 2 . 4G2 ) and were double stained , simultaneously , with anti-mouse surface CD4 ( L3T4 ) and CD8 ( 53-6 . 7 ) conjugated to FITC and Cy-Chrome , respectively . For intracellular staining of cytokines , cells were permeabilized using Cytofix/Cytoperm ( BD Biosciences ) and were incubated with the anti-cytokine antibodies conjugated to PE: IFN-γ ( XMG1 . 2 ) , IL-4 ( BVD4-1D11 ) , and IL-10 ( JES5-16E3 ) . The isotype controls used were rat IgG2b ( A95-1 ) and rat IgG2a ( R35-95 ) . Data were collected and analyzed using CELLQuest software and a FACSort flow cytometer ( Becton Dickinson Immunocytometry System ) . The steady-state frequencies of cytokine positive cells were determined using lymph node cells from PBS-inoculated mice . L . braziliensis promastigotes ( strain MHOM/BR/01/BA788 [20] ) were grown in Schneider medium ( Sigma ) supplemented with 100 U/ml of penicillin , 100 ug/ml of streptomycin , 10% heat-inactivated fetal calf serum ( all from Invitrogen ) and 2% sterile human urine . Two weeks following the last immunization with L . intermedia SGS , mice were challenged in the left ear dermis by inoculation of stationary-phase promastigotes ( 105 parasites in 10 µl of saline ) +SGS ( equivalent to 1 pair of salivary glands ) . Lesion size was monitored weekly , for 16 weeks , using a digital calliper ( Thomas Scientific ) . Parasite load was determined using a quantitative limiting-dilution assay as described [22] . Briefly , infected ears and retromaxillar draining lymph nodes ( LNs ) were aseptically excised at two , four and 16 weeks post infection and homogenized in Schneider medium ( Sigma ) . The homogenates were serially diluted in Schneider medium supplemented as before and seeded into 96-well plates containing biphasic blood agar ( Novy-Nicolle-McNeal ) medium . The number of viable parasites was determined from the highest dilution at which promastigotes could be grown out after up to two weeks of incubation at 25°C . For measurement of in vitro cytokine production , single-cell suspensions of draining lymph nodes were prepared aseptically at two , four and 16 weeks post infection . The cells were diluted to 5×106 cells/ml in RPMI 1640 supplemented with 2 mM L-glutamine , 100 U/ml of penicillin , 100 ug/ml of streptomycin , 10% fetal calf serum ( all from Invitrogen ) and 0 . 05 M 2-ME and dispensed into 96-well plates with L . braziliensis ( ratio 1∶5 , parasite:cells ) +L . intermedia SGS ( equivalent to 2 . 5 pair of salivary glands/ml ) or L . intermedia SGS ( equivalent to 2 . 5 pair of salivary glands/ml ) alone . Cultures were incubated at 37°C in 5% CO2 . Supernatants were harvested at 48 h and assayed for IL-4 or at 72 h and assayed for IFN-γ . Cytokine presence was determined by ELISA using commercial kits ( BD Biosciences ) . Human sera used in the present study were obtained from an epidemiological survey conducted in a region endemic for American cutaneous leishmaniasis ( Canoa , a rural village , located near Santo Amaro , Bahia , Brazil ) . Details of the area , patients and anti-Leishmania ( Delayed Type Hypersensitivity ) DTH skin test are described elsewhere [23] . Informed consent was obtained from patients or their guardians and all procedures were approved by the local Ethics Committee ( CEP/MCO/UFA ) and were conducted following recommendations outlined in the Helsinki Declaration . Briefly , after an outbreak of cutaneous leishmaniasis ( CL ) in Canoa , three groups of individuals were characterized: 1 ) Patients who developed CL; 2 ) Individuals who converted the DTH skin test to Leishmania antigen and who did not develop disease and 3 ) Individuals who had a negative anti-Leishmania DTH skin test . ELISA for detection of anti-saliva antibodies in these three groups was performed as described in [24] . Control sera were obtained from individuals residing in Salvador , BA , an area not endemic for leishmaniasis . These individuals showed negative responses to both anti-Leishmania sorology and anti-Leishmania DTH skin test . ELISA was performed as above . Briefly , microplates were coated with SGS diluted to 5 pairs of salivary glands/ml in coating buffer , wells were blocked , incubated overnight with human sera ( 1∶50 dilution ) , followed by alkaline phosphatase-conjugated anti-human IgG ( 1∶5000 dilution ) . Data are presented as mean±standard error of the mean . The significance of the results was calculated by Mann Whitney or Kruskal-Wallis tests using Prism ( Graph Pad Software ) and P-values <0 . 05 were considered significant . To evaluate disease burden in mice , ear thickness of mice immunized with SGS and challenged with SGS+L . braziliensis was recorded weekly , for each individual mouse . The course of disease for experimental and control mice was plotted individually and the area under each resulting curve was calculated using Prism ( Graph Pad Software ) . The significance of the results ( area under curve obtained for each mouse immunized with SGS versus area under curve obtained for each mouse immunized with PBS ) was calculated by Mann Whitney . In studies performed with human sera , significance of the results was calculated by Kruskal Wallis test followed by Bonferroni's Multiple Comparison Test .
Sera from immunized mice obtained two weeks after the last immunization were able to recognize L . intermedia SGS by ELISA ( Fig . 1A ) and , importantly , the presence of IgG1 was significantly higher in mice L . intermedia SGS-immunized when compared to PBS-immunized mice . Similar differences were not observed when we compared IgG2a and IgG2b levels in SGS and PBS-immunized mice ( Fig . 1B ) . Western blot confirmed that immune sera specifically recognized the majority of bands seen in L . intermedia SDS-PAGE profiles ( Fig . 1D and 1C , respectively ) . When L . intermedia SGS immune sera were probed against L . longipalpis SGS , the only protein recognized was a protein of approximately 45 kDa , a protein with similar mobility to the yellow related protein from L . longipalpis ( Fig . 1D ) . We can speculate that , although the salivary gland protein profiles of both sand fly species appear similar ( Fig . 1C ) , their antigenic properties are different ( Fig . 1D ) . A mixed cytokine response , with the presence of both IFN-γ and IL-4 was observed in immunized animals ( Fig . 2A ) . These cytokines were detected within CD4+ ( left panel ) and CD8+ T cells ( right panel ) , at comparable levels . IL-10 expression was also detected in these two cell populations , albeit in a lower frequency when compared to IFN-γ and IL-4 . L . intermedia SGS also induced an important inflammatory response in immunized mice . Injection of SGS following immunization with SGS elicited , at 24 h post SGS injection , an inflammatory infiltrate comprised of numerous polymorphonuclears ( PMNs ) and few mononuclear cells ( Fig . 3A and B ) . Edema and vascular congestion were also noted . At 48 h post challenge injection with SGS , the infiltrate was more pronounced , with higher numbers of PMNs and mononuclear cells and accentuated myositis ( not shown ) . In mice which received PBS and were later challenged with SGS , there was edema formation with rare inflammatory cells at 24 h ( Fig . 3C and D ) . This effect was decreased at 48 h ( data not shown ) . We then examined whether immunization with L . intermedia saliva altered the course of L . braziliensis infection in BALB/c mice , using an intradermal experimental model of infection [20] . Surprisingly , pre-immunization with L . intermedia SGS did not protect against lesion development . As shown in Fig . 4A , the onset of lesion development in PBS-inoculated mice was at three weeks post infection , peaking at five weeks and resolving , spontaneously , by returning to normal ear thickness after ten weeks . In L . intermedia SGS immunized mice , however , dermal lesions appeared later , at five weeks post infection and peaked at eight weeks post infection ( Fig . 4A ) . From then on , lesions slowly declined and resolved at 16 weeks post infection , although ear thickness never returned to normal levels ( 0 . 2–0 . 3 mm ) . No significant differences in ear thickness , between both groups , were observed throughout infection . However , two weeks following infection , SGS–immunized mice showed a slight inflammatory infiltrate with a discrete increase in PMNs ( data not shown ) . Importantly , a significant difference ( p<0 . 05 ) was observed when we evaluated disease burden in SGS and PBS-immunized mice ( Fig . 4B ) . Disease burden was calculated by weekly measure of ear thickness and by comparison of the area under the resulting curves , as explained under statistical analysis . As shown in Fig . 4B , dermal lesions persisted for a longer period of time in SGS-immunized mice indicating that disease burden is more pronounced in these mice when compared to PBS-immunized mice . We then examined whether there was a correlation between lesion development and parasite replication in L . intermedia saliva immunized mice . As shown in Fig . 5 , two weeks after challenge , parasite load in the ear dermis and draining lymph nodes was significantly lower in saliva immunized mice ( Fig . 5A and B ) . Four weeks after challenge , parasite load was similar in both groups of animals , in the two compartments analyzed , which correlated with the similar lesion sizes ( Fig . 4A ) . Later , 16 weeks following challenge , mice immunized with L . intermedia SGS displayed a significantly higher parasite load in the ear ( Fig . 5A ) whereas no significant differences were observed in the draining lymph nodes of both groups ( Fig . 5B ) . At this same time point , however , there was no significant difference in ear thickness in both groups of mice . Next , we examined the immune response in mice immunized with L . intermedia saliva and challenged with L . braziliensis plus SGS . Two weeks following infection , the IFN-γ to IL-4 ratio is two-fold higher in animals inoculated with PBS when compared to mice immunized with SGS ( Fig . 6 ) . This higher IFN-γ production in relation to IL-4 was observed following in vitro stimulation with either L . braziliensis+SGS ( Fig . 6A ) or SGS alone ( Fig . 6B ) . Later on , the IFN-γ to IL-4 ratio in PBS inoculated mice , decreased steadily , correlating with the decrease in ear thickness ( Fig . 4A ) and in parasitemia at the inoculation site ( Fig . 5A ) . Overall , SGS immunized mice produced much less IFN-γ in relation to IL-4 after stimulation with either Leishmania plus SGS or after stimulation with SGS alone , which can be correlated with the higher disease burden ( Fig . 4B ) and the constant increase in parasite load ( Fig . 5A ) . Data obtained in mice immunized with L . intermedia saliva and later challenged with L . braziliensis showed that , in this case , saliva immunization failed to protect against a challenge infection with L . braziliensis . Although immunization with L . intermedia saliva delayed lesion appearance , lesions persisted for longer periods when compared to PBS inoculated mice . Upon this finding , we then asked whether human antibody response to L . intermedia saliva could be used to monitor exposure to sand fly and , possibly , be used as a marker of disease . In order to do so , we first investigated the anti-L . intermedia humoral immune response in individuals living in an endemic area for cutaneous leishmaniasis ( CL ) . As shown in Fig . 7A , CL endemic area individuals possessed significantly higher anti-L . intermedia SGS IgG levels when compared to control sera ( sera obtained from individuals residing in a non-endemic area ) . When sera from a CL endemic area were probed against L . longipalpis SGS , IgG levels were significantly lower when compared to IgG levels against L . intermedia SGS , illustrating the specificity of the anti-saliva immune response . We then selected , among individuals from the endemic area with positive anti-L . intermedia SGS IgG response ( Fig . 7A ) , a subgroup of patients with active lesion at the time of serum collection ( CL ) , a subgroup of individuals with positive anti-Leishmania DTH skin test and a subgroup with negative DTH skin test . Sera from these three subgroups were again probed against L . intermedia SGS . As shown in Fig . 7B , CL patients displayed a significantly higher IgG immune response against saliva when compared to individuals with positive DTH . This was also observed when comparing CL patients and DTH negative individuals . These results show that humoral immune response to L . intermedia SGS is a marker of disease in CL . Western blot analysis ( Fig . 8 ) showed that sera from CL individuals displayed the strongest response against L . intermedia SGS , when compared to DTH− and DTH+ individuals . This strong response was observed particularly for the 62 kD , 49 kD , 45 kD , 36 Kd , 28 kD and 14 kD bands . Sera from DTH− individuals , on the other hand , recognized , preferentially , the 28 kD and , in some cases , the 14 kD and 36 kD bands . Last , sera from DTH+ individuals showed the weakest response , when compared to CL and DTH− individuals . Nonetheless , these sera were able to detect , for some individuals , the 45 kD , 36 kD and 28 kD bands . Sera from controls individuals ( - ) did not recognize any of the salivary proteins .
The results presented herein show that L . intermedia SGS shifted the immune responses to L . braziliensis to type 2 and immunization of BALB/c mice with SGS led to enhanced L . braziliensis infection . Additionally , induction of this distinctive role played by L . intermedia saliva comes from data obtained with patients with active CL who presented higher anti-L . intermedia SGS antibody titers when compared to exposed individuals with positive anti-Leishmania DTH . Initially , we investigated the immunogenic properties of L . intermedia , the main vector of L . braziliensis in Brazil . We observed that mice immunized with L . intermedia SGS developed a specific humoral immune response as shown by ELISA and Western blot . The main subclass present in immune sera was IgG1 . In the absence of IgG2a , this is indicative of a preferential type 2 immune response , knowingly associated with susceptibility to leishmaniasis ( revised in [25] . The presence of IgG1 was also observed upon natural exposure of mice to L . longipalpis bites [5] and upon immunization with P . ariasi SGS [26] . In an area endemic for visceral leishmaniasis , exposed individuals showed a mixed composition of IgG1 and IgE antibodies to L . longipalpis saliva , suggesting a mixed type1/type 2 response [27] . This type of response has also been detected herein as shown by the presence of both IFN-γ and IL-4 in cell culture supernatants from L . intermedia SGS immunized mice . Specificity of the humoral immune response was tested by probing immune sera with L . longipalpis SGS , this species being the vector of L . chagasi , the causative agent of American visceral leishmaniasis . Interestingly , SDS-PAGE profiles of salivary proteins from both sand flies showed bands migrating at similar molecular weights . However , L . intermedia immune sera did not recognize any proteins present in L . longipalpis SGS with the exception of the ∼45 kDa protein . The species-specificity of sand fly salivary gland components has been examined elsewhere [28] , [29] . Thiakaki et al . [30] showed that mice exposed to bites of three sand fly species developed antibodies specific to the different saliva antigens , indicating the specificity of anti-saliva immune responses . Indeed , results from our laboratory have shown that salivary content similarity between L . intermedia and L . longipalpis is low ( de Oliveira et al . , manuscript in preparation ) corroborating the lack of antibody of cross-reactivity and emphasizing the uniqueness of the sand fly saliva antigens . Surprisingly , immunization with L . intermedia SGS did not confer any protection against L . braziliensis development: SGS immunized mice developed larger lesions which were maintained for a longer period when compared to PBS inoculated mice . As observed before in L . braziliensis infected mice [20] , parasites were able to persist in draining lymph nodes , regardless of lesion healing . The lack of protection against a challenge infection in SGS immunized mice can therefore be correlated with the low IFN-γ to IL-4 ratio observed throughout the infection period , an effect not observed in PBS inoculated mice . In the latter , disease burden was significantly lower , indicating their ability to mount and sustain a balanced immune response . Initially , SGS immunized mice showed a significantly lower parasite burden after challenge with parasites plus saliva . It is possible that this early control in parasitemia may be exerted by inflammatory cells ( mono and polymorphonuclear cells ) that are recruited following stimulation with saliva . This early control , however is not maintained since parasite multiplication was clearly observed , probably as a result of the pathogen-favorable adaptative immune response that is developed in SGS immunized mice . It has been shown that macrophages phagocytose apoptotic neutrophils infected with L . major [31] and also that coinjection of dead neutrophils amplified L . major replication in vivo in BALB [32] . It is tempting to speculate that the rise in parasite burden observed later on is related to similar effects exerted by the PMNs recruited following immunization with L . intermedia saliva and this hypothesis is under current investigation . In the L . major model , where immunization with P . papatasi SGS leads to protection against challenge , it is postulated that SGS immunization precludes the production of type 2 cytokines [8] , [12] . It was also shown that immunization with SP15 , antigen present in P . papatasi saliva , leads to the development of an anti-saliva DTH response [14] , which was also associated with protection . Herein , immunization with L . intermedia SGS did not inhibit the production of type 2 cytokines or promoted the development of a classical DTH reaction , with the characteristic presence of macrophages and lymphocytes . Accordingly , immunization with L . intermedia SGS is favoring L . braziliensis establishment and persistence in BALB/c mice . In human visceral leishmaniasis , individuals with positive anti-Leishmania cellular immune response ( a putative marker of protection against L . chagasi ) have increased anti L . longipalpis saliva IgG levels when compared to individuals with positive anti-Leishmania humoral immune response ( a marker of L . chagasi infection ) [24] , [27] . In the present study , however , we found an opposite correlation: a higher anti-L . intermedia saliva immune response was observed in CL patients whereas individuals from the endemic area with a positive DTH skin test against Leishmania showed a lower IgG response anti saliva . It is tempting to speculate that the detrimental effects of sensitization with L . intermedia saliva observed in mice have a parallel in individuals exposed in the endemic area . Rohousosva et al . , [29] observed that L . tropica patients possessed significantly higher anti-P . sergenti saliva IgG levels when compared to healthy individuals from the same place . In this work , authors suggested that higher anti-saliva IgG levels may reflect a more frequent exposure to vector bites and , therefore , higher probability of L . tropica transmission . In our study , we also observed a higher anti-L . intermedia SGS humoral response in individuals from a CL endemic area when compared to control individuals which may also reflect a more frequent exposure to vector bites . More importantly , however , we found that , among individuals from the endemic area , anti-saliva IgG responses were significantly lower in both DTH+ ( individuals who developed an anti-Leishmania cellular immune response when exposed to infected sand flies ) and in DTH− ( individuals who did not develop an anti-Leishmania cellular immune response ) , when compared to the anti-saliva response in CL patients . It would be most interesting to determine whether DTH− individuals either convert to DTH+ or develop CL and whether the anti-L . intermedia saliva humoral response , in these cases , increases or decreases , respectively . Nonetheless , the current results enable us to hypothesize that , for CL caused by L . braziliensis , a higher anti-saliva humoral immune response could be used as a marker of risk for Leishmania transmission , a finding reported for the first time in New World cutaneous leishmaniasis . Characterization of the immune response to sand fly saliva has proven useful in understanding how the molecules present therein modulate host's immune response . In the case of P . papatasi and L . major , a vaccine candidate has been identified and this type of study now constitutes a major area of research in the development of control measures against leishmaniasis [26] . However , we have shown that this is not necessarily applicable to other vector/parasite systems such as L . braziliensis/L . intermedia . Our study was conducted with inoculation of SGS and we cannot , presently , exclude that the effects observed in the mouse model are due to an immune response against saliva alone and not against a mixture of saliva and structural components of the salivary gland or even LPS . Regarding the latter , L . intermedia SGS preparations were unable to induce TNF-α production following stimulation of human monocytes ( data not shown ) and such effect is probably due to the very low concentrations of LPS ( <0 . 1 ng/ml , below the detection limit of the Limulus amebocyte assay ) . Indeed , it would be very interesting to determine the outcome of L . braziliensis infection in mice sensitized to L . intermedia sand fly bites . However , rearing of L . intermedia sand flies under laboratory conditions has proven a challenging process . To date , there has been very little work on transmission with L . braziliensis parasites , especially using the sand fly vectors , and experimental transmission of Viannia parasites by infected sand fly bite has not been achieved [33] . However , it has been shown that , regarding the use of SGS , mice develop a strong DTH response after a double exposure to the equivalent of 0 . 2 pairs of salivary glands , inoculated intradermally in the ear , and that a similar cellular infiltrate is mobilized to the skin following sand fly bites [34] . Moreover , it was also shown that there is little difference in the composition of sand fly saliva and SGS since major protein components of SGS are present in the salivary contents [35] , confirming that SGS is an appropriate source of antigen to mimic natural exposure to sand fly saliva . Nonetheless , since vector based vaccines are an important alternative envisaging disease control , we are currently investigating whether immunization with individual L . intermedia salivary components is capable of inducing protection . | Parasites of the genus Leishmania cause a variety of diseases known as leishmaniasis , that are transmitted by bites of female sand flies that , during blood-feeding , inject humans with parasites and saliva . It was shown that , in mice , immunity to sand-fly saliva is able to protect against the development of leishmaniasis . We have investigated , in the present study , whether this finding extends the sand fly species Lutzomyia intermedia , which is responsible for transmission of Leishmania braziliensis , a parasite species able to cause destructive skin lesions that can be fatal if left untreated . We observed that mice injected with sand fly saliva develop a specific immune response against salivary proteins . Most importantly , however , this immune response was unable to protect mice against a challenge infection with L . braziliensis , indicating that exposure to this sand fly saliva is harmful to the host . Indeed , subjects with cutaneous leishmaniasis have a higher immune response against L . intermedia saliva . These findings indicate that the anti-saliva immune response to sand fly saliva plays an important role in the outcome of leishmaniasis caused by L . braziliensis , in both mice and humans , and emphasize possible hurdles in the development of vaccines based on sand fly saliva . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"immunology/immunomodulation",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"immunology/immune",
"response",
"infectious",
"diseases/protozoal",
"infections",
"immunology/immunity",
"to",
"infections"
] | 2007 | Enhanced Leishmania braziliensis Infection Following Pre-Exposure to Sandfly Saliva |
Helminth co-infection in humans is common in tropical regions of the world where transmission of soil-transmitted helminths such as Ascaris lumbricoides , Trichuris trichiura , and the hookworms Necator americanus and Ancylostoma duodenale as well as other helminths such as Schistosoma mansoni often occur simultaneously . We investigated whether co-infection with another helminth ( s ) altered the human immune response to crude antigen extracts from either different stages of N . americanus infection ( infective third stage or adult ) or different crude antigen extract preparations ( adult somatic and adult excretory/secretory ) . Using these antigens , we compared the cellular and humoral immune responses of individuals mono-infected with hookworm ( N . americanus ) and individuals co-infected with hookworm and other helminth infections , namely co-infection with either A . lumbricoides , Schistosoma mansoni , or both . Immunological variables were compared between hookworm infection group ( mono- versus co-infected ) by bootstrap , and principal component analysis ( PCA ) was used as a data reduction method . Contrary to several animal studies of helminth co-infection , we found that co-infected individuals had a further downmodulated Th1 cytokine response ( e . g . , reduced INF-γ ) , accompanied by a significant increase in the hookworm-specific humoral immune response ( e . g . higher levels of IgE or IgG4 to crude antigen extracts ) compared with mono- infected individuals . Neither of these changes was associated with a reduction of hookworm infection intensity in helminth co-infected individuals . From the standpoint of hookworm vaccine development , these results are relevant; i . e . , the specific immune response to hookworm vaccine antigens might be altered by infection with another helminth .
Helminth co-infection in humans is common in tropical regions [1] , [2] , where transmission of Ascaris lumbricoides , Trichuris trichiura , the hookworms ( N . americanus or A . duodenale ) , and schistosomes often occur concurrently [3] , [4] . Although co-infection is often the rule rather than the exception in endemic areas , most previous immuno-epidemiological studies of human helminth infection have focused on the immune response to a single helminth species ( mono-infection ) rather than the more common situation where an individual is infected with one or more different helminth species [5] . At our study site in Northeastern Minas Gerais State , Brazil , where co-infection with schistosomes and soil-transmitted helminths ( STHs ) is common [6] , we have attempted to study the epidemiologic , immunologic , and genetic determinants of infection in individuals resident in these co-endemic areas [7]–[11] . Much of the previous information on the immunology of helminth co-infections has come from laboratory animal models , especially experimental rodent models . The majority of these studies show a competition between the co-infections , with one infection usually leading to the rapid expulsion of the other [12]–[16] . The immune mechanisms behind this effect are hypothesized to include cross-reactive antibodies ( also referred to as “cross-protection” ) [13] , [16] , a skewing towards Th2 cytokines ( e . g . , elevated IL-4 ) , increased Th2-type antibody isotypes ( e . g . , elevated production of IgG1 ) [15] , and mucosal mast cell activation [13]–[15] . However , conflicting animal studies report that co-infection increases infection intensities by down modulating Th2 cytokine responses , which in turn reduces intestinal inflammation , leading to slower worm expulsion and increased worm burdens in co-infected animals [17] . Possible explanations for these opposite findings , among others , might be differences in animal models , different combinations of parasite infections , and the different timing of co-infection ( by timing of the primary versus the secondary infection ) . The few studies on the human immune response in co-infected individuals are also contradictory . In one group of studies , helminth co-infection appeared to result in a synergistic effect among the infections , with infection with one helminth being associated with an increased risk of having a high intensity infection with another helminth [7] . However , other studies imply a cross-protective effect derived from co-infection: for example , individuals mono-infected with hookworm or A . lumbricoides develop antibodies that cross-react with antigens from S . mansoni [18]–[20] . In another set of studies , co-infection appeared to skew the immune response away from the helminth infection under study , e . g . , the humoral and cellular immune responses to hookworm or Ascaris antigens are diminished in individuals resident in a schistosomiasis endemic area [21] . Along these same lines , studies have also demonstrated an upregulation of the immune response during helminth co-infection; e . g . , increased production of inflammation markers to S . mansoni infection in children who are also infected with hookworms and/or Entamoeba species [22] . However , given the contradictory nature of these outcomes , the central question of whether multiple helminth infections drive host immune responses towards phenotypes different from those of a single infection still remains to be answered [23] . In our previous epidemiological study in Brazil , we showed synergistic effects among helminth co-infections in terms of egg counts [7] , leading us to expect a similar synergistic effect on immune responses during helminth co-infection . In keeping with the results from experimental animal studies [12]–[16] , we further hypothesized that hookworm co-infections with A . lumbricoides and/or S . mansoni would significantly alter the immune responses to crude hookworm antigen extracts , resulting in reduced Th2-type responses ( IL-4 , IL-5 , IL-13 ) , a reduced inflammatory response ( e . g . , lower TNF-α secretion ) , and an increase in the production of regulatory cytokines ( e . g . , IL-10 ) . To test this hypothesis , we compared the cellular and humoral immune responses of individuals infected with hookworm alone ( mono-infected ) and individuals infected with hookworm and either A . lumbricoides , S . mansoni or both ( co-infected ) .
The study was conducted in an area of the northeastern part of the state of Minas Gerais in Brazil that is endemic for S . mansoni and the STH as previously described [7] . The area of Americaninhas is divided into five rural sectors and a central municipality . The Fundação National de Saúde ( the National Health Foundation ) estimates the population to be approximately 1000 in the urban municipal center and another 1000 in the surrounding rural areas . Each house was assigned a unique household identification number ( HHID ) , and each resident , a unique personal identity number ( PID ) . Only individuals meeting the following inclusion criteria were included into the study: ( 1 ) resident in the study area over the last 24 months; ( 2 ) reporting not to have received anthelmintic treatment within the last 24 months; and ( 3 ) willing and able to give informed consent to study protocol . Individuals were not included if they: ( 1 ) attended school outside the study area; ( 2 ) worked full-time outside the study area; or ( 3 ) tested positive on a pregnancy test . Females found to be pregnant during the test were excluded from treatment during their pregnancy and received treatment for all helminth infections later . For parasitological exams , participants were instructed to deposit one fecal sample per day into each container and return the container to one of several collection points , where the sample was stored at 4°C . Fecal samples returned later than 48 h after date of distribution were not accepted , and new containers were issued . Presence of infection was determined by using the formalin-ether sedimentation technique . Individuals positive for any helminth in the formalin-ether sedimentation technique were asked to contribute two more samples over the course of two more days to be analyzed by Kato-Katz technique for assessment of eggs per gram of feces ( infection intensity ) . Two slides were taken from each day's fecal sample for a total of four slides from each individual . Slides were examined within 45 minutes of slide preparation to avoid drying of hookworm eggs . The arithmetic means of the four slides was calculated and then converted to eggs per gram according to the Kato-Katz method [24] . Out of 1 , 332 consented participants in the study , two-hundred and fifty individuals were selected by simple random sampling for immunological assays . Random sampling was performed on an age , gender , and infection stratified sampling frame . In brief , individuals with a negative fecal exam were removed from the sampling frame; i . e . , only persons with a positive fecal exam were included . The sampling frame was then divided into 10 mutually exclusive and exhaustive gender-based strata based using the following age intervals: <9 , 10–19 , 20–29 , 30–39 , and >40 years of age . Simple random sampling was performed independently in each stratum . Individuals who refused to enroll in this part of the study or who were not eligible were replaced by simple random sampling from the same stratum . The final stratified random sample was compared to non-participants for age , gender , and infection intensity , and no statistically significant differences ( p>0 . 05 ) were found in terms of those variables between those individuals included in the survey and those not . Individuals found to be infected with hookworm or other intestinal nematodes were treated with albendazole ( 400 mg ) . Participants with schistosomiasis were treated with praziquantel ( 50 mg/kg ) under the supervision of the project physician . In the present study , cellular and humoral immune responses from individuals with a hookworm mono-infection [9] were included , as well as from individuals co-infected with ( a ) hookworm and A . lumbricoides , ( b ) hookworm and S . mansoni , or ( c ) hookworm , A . lumbricoides and S . mansoni . After parasitological exams and before anthelminthic treatment , approximately 20 mL of blood was collected in heparinized tubes from children ≥6 years of age and adults for separation of peripheral blood mononuclear cells ( PBMC ) and 4 mL of blood in EDTA tubes for the immunological assays described below . The study was approved by the ethical review committees of The George Washington University ( GWU , USA ) , the London School of Hygiene and Tropical Medicine ( UK ) , the Centro de Pesquisas René Rachou FIOCRUZ and the Brazilian National Committee for Ethics in Research ( CONEP ) , and all subjects provided written informed consent to participate in the study , or , in the case of minors , written informed consent was given by their parents or guardians . Phenotyping of lymphocytes was performed as described elsewhere [9] and the following pairs of monoclonal antibodies ( mAb ) , either conjugated with phycoerythrin ( PE ) or fluorescein isothiocyanate ( FITC ) were used: CD4 ( FITC ) /CD25 ( PE ) , CD4 ( FITC ) /HLA-DR ( PE ) , CD4 ( FITC ) /CD45RO ( PE ) , CD4 ( FITC ) /CD45RA ( PE ) , CD8 ( FITC ) /CD28 ( PE ) , CD8 ( FITC ) /HLA-DR ( PE ) , CD8 ( FITC ) /CD45RO ( PE ) , CD8 ( FITC ) /CD45RA ( PE ) , CD3 ( FITC ) /CD69 ( PE ) , and CD19 ( FITC ) /CD27 ( PE ) . Mouse IgG1 antibodies conjugated with FITC or PE served as isotype controls . Sample acquisition was done on a FACScan flow cytometer ( Becton Dickinson , USA ) and results for 10 , 000 events were analysed with BD Cell Quest™ software ( Becton Dickinson , USA ) . For the evaluation of humoral and cellular immune responses , soluble somatic antigen extracts were prepared from third-stage larvae ( L3 ) and adult worms ( AE ) of Ancylostoma caninum . Excretory/secretory ( ES ) antigens were obtained from cultured A . caninum adult worms . The preparations were performed as described elsewhere [9] . For the detection of parasite-specific IgE antibodies , each of the hookworm antigens were diluted with carbonate buffer ( pH 9 . 6 ) to a concentration of 5 µg/ml . High-binding ELISA plates ( NUNC , Maxisorp , Fisher Scientific , USA ) were coated with 100 µl of the diluted antigens and incubated overnight at 4°C . Plates were washed 5 times with washing buffer ( phosphate buffered saline [PBS]/0 . 05% Tween-20; pH 7 . 2–7 . 4 ) and were then blocked for 1 hour at room temperature ( RT ) with 200 µl of blocking buffer ( PBS/ 0 . 05% Tween-20/ 3% bovine serum albumin ) . Individual serum samples were diluted 1∶50 in blocking buffer , 200 µl were added in duplicate to the respective wells , and plates were incubated overnight at 4°C . On the following day , plates were washed 10 times with washing buffer . A 1∶1 , 000 dilution of anti-human IgE alkaline phosphatase-conjugated antibody ( Pharmingen , USA ) was prepared in PBS/0 . 05% Tween-20 and 100 µl were added to the wells . After another incubation of 90 minutes at RT , plates were washed 5 times and then 100 µl of p-nitrophenyl phosphate substrate was added to each well . Plates were incubated overnight at 4°C and the following morning the color reaction was read at 405 nm using an automated ELISA reader ( SpectraMax 340 PC , Molecular Devices , USA ) using SOFTmax Pro 5 . 2 for Windows ( Molecular Devices ) for data capture . Reference sera were assayed on each plate as positive and negative controls . For detection of parasite-specific IgG subclasses , horseradish peroxidase-conjugated , anti-human IgG1 , IgG3 , and IgG4 ( Zymed , USA ) were used at a dilution of 1∶1000 , as described above . As substrate , ortho-phenylene diamine was used and the color reaction was stopped with H2SO4 after incubation for 30 min at RT in the dark . Plates were read at 490 nm . The separation of lymphocytes , their stimulation in vitro with different hookworm antigens and with the mitogen phytohemagglutinin ( PHA ) , lymphocyte proliferation , as well as the secretion of several cytokines and chemokines after in vitro stimulation were performed as described elsewhere in detail [9] . Here we report the proliferation of lymphocytes after stimulation with the crude soluble hookworm antigens L3 , AE , and ES . For in vitro cytokine or chemokine secretion , lymphocyte cultures were stimulated with the same antigens and with PHA , as described for proliferation assays , and the following analytes were measured: Interleukin ( IL ) -2 , IL-4 , IL-5 , IL-10 , IL-13 , CXCL10 , TNF-α , and IFN-γ . The intensity of hookworm infection ( as determined by fecal egg counts ) was compared between groups by non-parametric Kruskal-Wallis test . Associations between Necator intensity of infection and antibody level against crude antigen extracts or Necator infection intensity and secreted cytokines/chemokines were analysed by Spearman's rank correlation . Analyses of these immune responses were done separately for the different co-infection combinations and then compared with hookworm mono-infected individuals . As the results among the different co-infection subgroups were found to be generally similar ( see below , in particular Table 1 and Figure 1 ) , we merged the various co-infections into a single group . For the chemokine and cytokine variables , analysis was done on the log-transformed variables , after replacing any zero values with 1 . Immunological variables were compared by bootstrapping the geometric mean after adjusting for age by linear regression on the log-values . For the lymphocyte populations , the untransformed values were used and hence the arithmetic means were compared . The immunological variables were summarized using principal component analysis ( PCA ) , via a projection-pursuit algorithm robust to departures of the data from normality [25] , [26] . We then used biplots [27] to simultaneously show i ) the contributions of each of the original variables to the first two principal components ( the ‘loadings’ ) , and ii ) each person's value of the principal components ( the ‘scores’ ) . The bivariate score means and their 95% confidence ellipses [28] were calculated for the mono-infected and co-infected groups . These means were compared between infection groups by the multivariate Hotelling's T2 test [29] . PCA analysis was done for lymphocyte sub-populations , for antibody responses , and for chemokine and cytokine response to three hookworm antigen preparations ( AE , ES , and L3 ) and a mitogen ( PHA ) Pairs of correlation coefficients by infection group were compared by first transforming the variable to a standard normal deviate via the Fisher Z transformation . No adjustment for multiple comparisons was made in these analsyses . Analyses were performed using S-PLUS version 6 . 2 or later ( Insightful Corp , Seattle WA , USA ) and R version 2 . 10 or later ( R Foundation for Statistical Computing , Vienna , Austria ) . The PCA analysis used the ‘pcaPP’ package in R .
We observed a statistically significant increase in CD4/HLA-DR and CD8/HLA-DR positive T-cells in co-infected individuals compared to mono-infected individuals . Other comparisons of surface markers on T and B cells between mono- and co-infected individuals were not significant ( see Table 2 ) . PCA was performed on these immunological parameters jointly in order to obtain a more complete and integrated picture of the immunological pattern and compare the weight of each parameter's contribution to the immune response . The first principal component ( PC 1 ) was dominated by a contrast between CD4+/CD25+ ( positive loading ) and CD8+/CD28− T cells ( negative loading ) . PC 2 is effectively an average of CD4/CD45RA and CD8/CD45RA positive memory T cells ( see Figure S1 ) . In participants either mono-infected or co-infected , we found positive correlations between individual fecal egg counts and serum IgG4 antibody levels against all the hookworm crude antigen preparations tested: L3 , AE and ES . Other isotypes , such as IgG1 , IgG3 , and IgE , were not strongly correlated with egg counts ( see Table S1 ) . For individuals with co-infections , the correlations between fecal hookworm egg counts and hookworm-specific IgG4 were significant for AE ( rho = 0 . 40; p<0 . 001 ) , ES ( rho = 0 . 21; p = 0 . 007 ) , and L3 ( rho = 0 . 26; p = 0 . 001 ) antigen preparations . Optical density values for hookworm-specific serum antibodies were measured and the age-adjusted ratio between mono- and co-infected individuals are shown in Table 3 , where we observed significantly higher values for L3-specific IgG3 , IgG4 , and IgE , AE-specific IgG1 , IgG4 , and IgE , and ES antigen specific IgG1 and IgG4 responses in co-infected individuals compared to mono-infected individuals ( Table 3 ) . Mean PC values for mono-infected and co-infected individuals , plus their 95% confidence intervals ( ellipses ) , showed distinct segregation between these infection groups , with the mono-infected individuals having lower values of PC 1 , which was dominated by IgG3 against AE antigen , and IgE against AE and ES antigens ( Figure 2 ) . PC 2 showed a contrast between i ) IgG1 and IgG3 against AE antigen ( positive loadings ) and ii ) IgE against AE and ES antigens ( negative loadings ) . Values for lymphocyte proliferation were indicated as stimulation indices , i . e . proliferation of antigen- or mitogen-stimulated cells divided by the proliferation of unstimulated control cultures . Analysis of lymphocyte proliferation did not result in any significant differences between mono- and co-infected groups ( data not shown ) . Non-parametric correlations between individual PBMC secreted cytokine or chemokine levels and fecal hookworm egg counts were strongly negative for IL-10 in mono-infected participants and significantly different when compared with co-infected individuals , whether stimulated with L3 or AE ( p = 0 . 032 for both comparisons ) , or ES antigen ( p = 0 . 003 , Table 4 ) . Likewise , strong negative correlations were found for TNF-α in control cultures from mono-infected individuals or when cells were stimulated with ES , which were significantly different from the co-infected group ( p = 0 . 002 and p = 0 . 04 , respectively , Table 4 ) . In individuals with co-infection , significant negative correlations between egg counts and CXCL10 secretion were found in cell cultures stimulated with L3 ( p<0 . 05 ) or ES antigen ( p<0 . 01 ) , however without any significant differences when compared with mono-infected individuals . Analysis of cytokine and chemokine production in PBMC after stimulation with L3 antigen resulted in a significantly higher production of CXCL10 in mono-infected individuals ( Table 5 ) . Also , in PBMC stimulated either with AE or ES crude antigen extracts , significantly higher concentrations of TNF-α or IFN-γ were observed in mono-infected individuals when compared with the co-infected group ( Tables 6 and 7 ) . Examples of PCA for antigen-specific cytokine and chemokine secretion are shown in Figures S2 , S3 . For AE , as well as for ES antigen stimulation of PBMC , the highest loadings for PC1 and PC2 with the same directions were obtained for both Th1- and Th2-type cytokines or chemokines .
This is the first study to comprehensively examine the hookworm-specific humoral and cellular immune response in individuals who are co-infected with other helminths in an area of high hookworm transmission . This is also the first study to examine the effect of co-infection on the immune response to crude hookworm antigen extracts from different stages of hookworm development ( L3 , AE , ES ) . Moreover , these effects were analyzed in an epidemiologically well-characterized group of individuals , where the spatial , genetic and demographic aspects of hookworm infection and co-infection have been intensively studied [7] , [8] , [10] , [11] . Apart from non-parametric methods and comparisons of individual parameters , we also utilized principal component analysis for comparison of the immune responses to hookworm crude antigen extracts between mono- and co-infected individuals , enabling us to examine , and compare numerous mutually correlated immune variables in relation to the effects of mono- or co-infection status [30] . Our analyses showed that chronic co-infection with nematode and trematode species considerably alters the immune response to hookworm crude antigen extracts . Most interestingly , co-infection altered to a significant degree the antigen-induced secretion of inflammatory TNF-α and led to a further diminution of hookworm-specific IFN-γ and CXCL10 secretion , but did not alter production of IL-10 or the Type-2 cytokines , when compared to mono-infected individuals . In contrast to our previous study [9] , we found that the immune response to hookworm infection was increasingly modulated in co-infected individuals , an alteration that did not lead to expulsion of one parasite species as shown in experimental co-infections of mice with S . mansoni and Trichuris muris [15] . These findings are extremely relevant for successful planning of a hookworm vaccine currently under development [31] . In areas endemic for hookworm , such as the one studied , co-infections with other helminth species like A . lumbricoides and Schistosoma are common . Our results show that Type 1 immune responses to hookworm are significantly altered by such co-infections , which might have implications for hookworm vaccine development , with recent hookworm vaccines focused on inducing a Th1 response [32] in order avoid problems with hookworm induced IgE . The major emphasis of our immunological study was on T cells , i . e . , the proliferation of T cells , activation of T cell subpopulations , and secretion of Th1- and Th2-type cytokines and chemokines . Changes in CD4 and CD8 T cell counts , together with increased activation of these T cell subpopulations , have already been reported for helminth infections [33] . We add to this literature the finding that percentages of activated CD4+ and CD8+ T cells increased with co-infection . We speculate that multiply-infected individuals have higher percentages of activated CD4+ and CD8+ T cells due to ongoing higher antigenic stimulation of the immune system by different helminth species and cross-reactive antigens . This is supported by in vitro experiments on naïve human PBMC stimulated with soluble egg antigen from S . mansoni ( SEA ) , which showed an increase in the CD4+/HLA-DR+ cell population after in vitro priming and a further increase during recall responses [34] . Even though mean fecal egg counts in mono-infected patients were found to be in the range of those from co-infected individuals , correlations between hookworm egg counts and hookworm-specific IgG4 responses were stronger in co-infected patients , which might be attributed to the presence of antibodies that were cross-reactive with antigens from co-infecting helminth species [21] , [35] , [36] . Chronic infections with multiple helminth species might induce a stronger and ongoing antigenic stimulation of the host's immune system , which may lead to the expansion of antigen-specific B cells and the secretion of specific IgG4 antibodies , especially in co-infected individuals with increased hookworm infection . In support of this , a prior study with volunteers co-infected with hookworm , S . mansoni , and A . lumbricoides showed an increase in helminth antigen-specific total IgG antibodies when compared with the respective mono-infected groups [21] . In hookworm infections , the production of all antigen-specific IgG subclasses rises with ongoing infection [35] and hookworm-specific IgG4 has been proposed as a good marker for patent and chronic infections [35]–[37] . Analysis of cytokine and chemokine secretion patterns from mono-infected volunteers revealed no clear polarization into Th1 or Th2 type immune responses , but rather a mixed pattern [9] . Similar results were recently obtained for individuals co-infected with A . lumbricoides and T . trichiura [38] . However , in the co-infected group , we found a decreased TNF-α secretion , together with a further down-modulation of hookworm-specific IFN-γ production . Another study on co-infection detected elevated levels of pro-inflammatory cytokines and chemokines in co-infected children in response to S . mansoni adult worm antigen , whereas IFN-γ and IL-13 secretion patterns revealed no significant differences between individuals mono- and poly-infected with schistosomes , hookworm and Entamoeba species [22] . As opposed to A . lumbricoides and Trichuris trichiura co-infections [38] , we were neither able to detect a positive relationship between hookworm antigen-induced IL-10 secretion and intestinal worminess , nor to detect negative associations between IL-10 and Th1/Th2-type cytokines . These described differences might be due to the presence of different parasite species and also due to a mixture of intestinal and extra-intestinal parasites . Considerable antigen-induced IL-10 secretion has been described in individuals with hookworm infection [9] , [39] . In the current study , IL-10 levels correlated inversely with fecal egg counts in mono-infected hookworm patients especially in response to ES . This strong negative correlation was ablated in co-infected individuals , most probably because A . lumbricoides and S . mansoni infections induce production of IL-10 themselves [39] . Even though there was an unexpected negative correlation between parasite load and IL-10 secretion of lymphocytes , the antigen-induced IL-10 secretion was significantly associated with mono-infected individuals , indicating its importance in immune regulation during hookworm infection . This study has some important limitations . First , the cross-sectional study design , in which groups are compared from a single time point , does not allow causal inferences to be made . In addition , the small sample size may have limited our ability to detect small statistical differences between groups . Nor does the sample size allow for further stratification of the groups in order to explore other factors which may account for these differences . Age is likely to be among the most important of such confounding factors but was included as a covariate when testing for differences between groups . One positive aspect of the study design was the population-based sampling which should enhance the generalizability of the study . In summary , individuals co-infected with other helminth species presented with a significantly different immune response when compared with mono-infected participants . These changes included a stronger activation of CD4+ and CD8+ T cells , lower secretion of Type 1 cytokines , and increased levels of IgG4 and IgE antibodies against somatic hookworm antigens ( L3 and AE ) . Furthermore , positive correlations between egg counts and hookworm-specific IgG4 responses , as well as missing correlations between egg counts and regulatory ( IL-10 ) and inflammatory ( TNF-α ) cytokines in co-infected individuals . This modulation of hookworm-specific cellular and humoral immune responses by co-infection with other helminth species will be an important consideration during clinical trials for hookworm vaccine testing . Although vaccination is obviously not the same as natural infection , the immunogenicity of hookworm antigens in a vaccine might be altered and adversely affected by infections with parasites such as S . mansoni and A . lumbricoides . | Parasitic infections in humans are common in tropical regions and under bad housing and sanitation conditions multiple parasitic infections are the rule rather than the exception . For helminth infections , which are thought to affect almost a quarter of the world's population , most common combinations include soil-transmitted helminths , such as hookworm , roundworm , and whipworm , as well as extra-intestinal infections by schistosomes . In order to develop and test a hookworm vaccine in endemic areas , the understanding of the impact of multiple helminth infections ( co-infection ) on the immune response against hookworm in infected individuals is crucial . The authors report in their article , that several parameters of the cellular ( T cell markers , cytokines , chemokines ) and humoral immune response ( e . g . IgG4 and IgE antibodies ) against hookworm are significantly affected or modulated in individuals co-infected with hookworm , roundworm and/or schistosomes . These results imply that the immune response against components of a hookworm vaccine might be altered by previous contact with other helminth species in endemic areas . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"schistosomiasis",
"soil-transmitted",
"helminths",
"immunity",
"immunity",
"to",
"infections",
"immunology",
"biology",
"neglected",
"tropical",
"diseases",
"hookworm",
"immunomodulation"
] | 2011 | Necator americanus and Helminth Co-Infections: Further Down-Modulation of Hookworm-Specific Type 1 Immune Responses |
AKAP200 is a Drosophila melanogaster member of the “A Kinase Associated Protein” family of scaffolding proteins , known for their role in the spatial and temporal regulation of Protein Kinase A ( PKA ) in multiple signaling contexts . Here , we demonstrate an unexpected function of AKAP200 in promoting Notch protein stability . In Drosophila , AKAP200 loss-of-function ( LOF ) mutants show phenotypes that resemble Notch LOF defects , including eye patterning and sensory organ specification defects . Through genetic interactions , we demonstrate that AKAP200 interacts positively with Notch in both the eye and the thorax . We further show that AKAP200 is part of a physical complex with Notch . Biochemical studies reveal that AKAP200 stabilizes endogenous Notch protein , and that it limits ubiquitination of Notch . Specifically , our genetic and biochemical evidence indicates that AKAP200 protects Notch from the E3-ubiquitin ligase Cbl , which targets Notch to the lysosomal pathway . Indeed , we demonstrate that the effect of AKAP200 on Notch levels depends on the lysosome . Interestingly , this function of AKAP200 is fully independent of its role in PKA signaling and independent of its ability to bind PKA . Taken together , our data indicate that AKAP200 is a novel tissue specific posttranslational regulator of Notch , maintaining high Notch protein levels and thus promoting Notch signaling .
Signaling pathways are critically involved throughout embryonic development , as well as adult tissue function and homeostasis . Many of these pathways are highly conserved from invertebrates to humans , and were first discovered in Drosophila melanogaster , making it an ideal model system for identification and analysis of new pathway components . Notch signaling , is one such pathway , and is required for fundamental developmental processes including polarity , cell fate specification , tissue growth , stem cell maintenance , and organ patterning [1–5] . Moreover , misregulation of Notch signaling underlies several human diseases including various cancers highlighting the importance of Notch pathway regulation [1 , 3 , 6–8] . In Drosophila , Notch signaling is activated by the interaction of the ligands Delta and Serrate with the extracellular domains of the Notch receptor [9] . Ligand binding activates extracellular cleavage of Notch by ADAM/TACE metalloproteases [10] , followed by γ-secretase mediated cleavage [11] , which releases the Notch intracellular domain ( NICD ) [12] . The NICD , which is the signal transducing end of the protein , enters the nucleus , forms a complex with the transcription factor Suppressor of Hairless [Su ( H ) ]/CSL and activates target genes [5 , 13–15] . The same fundamental elements/mechanisms of the pathway are conserved in mammals [16 , 17] . Notch signaling is tempered by endocytosis of the receptor and degradation of NICD and these processes are essential to avoid hyperactivity [18 , 19] . Several studies have demonstrated proteasomal degradation of Notch . For example , dominant negative mutations of proteasomal subunits enhance Notch signaling in Drosophila [20] . Initial evidence for lysosomal degradation of Notch came from a study in skeletal myoblasts , the C2C12 cell line , where a role was demonstrated for c-Cbl ( Casitas B-lineage lymphoma , a proto-oncogene and E3 ubiquitin ligase ) in mono-ubiquitinating the endogenous transmembrane Notch1 and targeting it for lysosomal degradation [21] . Suppressor of Deltex [Su ( dx ) ]/Itch ( Drosophila/mouse ) and Sel10 have been shown to decrease Notch signaling in this context [22–25] . Mutations in Notch affect many developmental decisions in various Drosophila tissues [26 , 27] . For example , Notch signaling instructs specification of the eye field and initiation of eye development , as well as controlling growth and cell fate [28–31] . The interplay between Notch and Frizzled ( Fz ) /Planar Cell Polarity ( PCP ) signaling is critical for induction of specific photoreceptor ( PR ) subtypes [29 , 30 , 32–34] . In the developing eye disc , there is a Frizzled/PCP activity gradient that is highest at the dorso-ventral midline , termed the equator , and lowest at each pole [30 , 35 , 36] . Within each developing PR cluster , there are pairs of cells that are initially equivalent that then develop into photoreceptor 3 and 4 ( R3 and R4 ) . Within each pair , the cell that is closest to the midline adopts the R3 fate and upregulates the Notch ligand Delta , and neuralized and signals via Notch to its polar neighbor to adopt the R4 fate [30 , 36–38] . In a screen for novel regulators of PCP signaling in the Drosophila eye , we identified a scaffolding protein , A Kinase Anchoring Protein 200 ( AKAP200 ) [39] . AKAPs are a family of proteins responsible for the subcellular compartmentalization of Protein Kinase A ( PKA ) , which facilitate the spatial and temporal regulation of signaling [40–42] . Despite being structurally and sequentially diverse , the AKAP family of proteins show functional conservation amongst species [43] . AKAP200 is one member of the AKAP family of proteins and is expressed throughout all stages of Drosophila development . Alternative splicing produces two isoforms of AKAP200—the full length AKAP200-Long ( AKAP200-L ) , and the short isoform , AKAP200-Short ( AKAP200-S ) . AKAP200-S lacks the PKA-interaction domain and may therefore be limited to PKA- independent functions [44 , 45] . Here we provide evidence that AKAP200 is required for the regulation of Notch protein levels , via the lysosomal degradation pathway . AKAP200’s loss and gain-of-function ( LOF/GOF ) phenotypes are characteristic of misregulation of the Notch signaling pathway . AKAP200 LOF mutants display defects in cell fate specification manifested as loss of PRs in the eye and extra sensory bristles in the thorax , while AKAP200 overexpression causes wing vein defects and tissue overgrowth in the wing . Genetic interaction studies revealed that AKAP200 acts as a positive regulator of Notch signaling , as loss of AKAP200 suppresses Notch overexpression phenotypes in the eye and thorax . Consistent with this , we observe a decrease in overall Notch protein levels and increased ubiquitination in the AKAP200 mutant relative to wild type ( WT ) . Importantly , AKAP200’s effects on Notch are independent of PKA . However , we find that the suppression of Notch hyperactivity in AKAP200 mutant tissues is instead dependent on the E3 ubiquitin ligase Cbl and the lysosomal degradation pathway . Based on these data , we postulate a novel mechanism for the regulation of Notch levels , with AKAP200 preventing Cbl-mediated lysosomal degradation of Notch .
To identify novel genes involved in PCP-mediated photoreceptor specification , we performed a genetic screen for dominant modifiers of a gain-of-function ( GOF ) of the core PCP factors dgo and pk [39] . Overlapping deficiencies narrowed down a region on chromosome 2L that enhanced the dgo GOF PCP eye phenotypes ( S1A Fig ) . Further analysis using RNA interference ( IR ) against specific genes in this interval revealed that A Kinase Anchoring Protein 200 ( AKAP200 ) reproduced the interaction , implicating it as the gene responsible ( S1A and S1B Fig ) . To investigate AKAP200 functions , we first generated mutant alleles by excising the coding sequence of AKAP200 using flanking piggyBac/FRT insertions ( S1C Fig ) . This led to the isolation of two null alleles , AKAP200M30 and AKAP200M24 that were confirmed by PCR characterization ( S1D Fig; see [46] for method ) . These were lethal homozygous , or transheterozygous over a deficiency chromosome , with rare escapers ( S1G–S1J Fig ) . To analyze the loss-of-function ( LOF ) phenotypes we generated mutant clones via the Flp-FRT system [47] . In the eye , AKAP200M30 mutant clones displayed loss of or misspecification of photoreceptors ( PRs ) ( Fig 1B , 1C and 1F ) . These phenotypes were also consistently seen in escapers from different AKAP200 transheterozygous LOF genetic backgrounds and also mimicked the phenotypes seen with AKAP200-IR knock-downs ( S1G and S1H Fig; these were often quantitatively weaker than the null clones ) . A frequent defect was loss of R7 in mutant ommatidia ( Fig 1B , 1C and 1F ) , with R7 specification requiring both Notch and RTK ( Sev and Egfr ) signaling . To confirm this we analyzed developing eye discs with molecular markers ( Elav to stain all R-cells and Pros labeling R7 ) , which revealed partial photoreceptor specific loss of Pros staining in mutant eye discs ( S1I and S1I’ Fig ) . Together with the identification in the PCP screen , the observed eye phenotype was suggestive of a possible link to Notch function , with similarities to aspects of Notch LOF phenotypes [28 , 48–50] . Also as an interplay between Notch and Fz/PCP signaling is critical for R3/R4 specification and PCP patterning in the eye , regulators of either pathway were expected to be and were identified in the screen [39] . The AKAP200 LOF thorax phenotype [generated using ubx-Flp inducing clones in all imaginal discs at early larval stages [51]] revealed supernumerary scutellar bristles , and loss of or mispositioned microchaetae ( Fig 1E , red arrow marks supernumerary bristles and white arrow a bald patch representing loss of bristles , compare to wild-type area with evenly spaced and regular positioning of microchaetae bristles ) . Strikingly , the supernumerary scutellar bristles ( macrochaetae; Fig 1E , red arrow ) are a hallmark Notch-/+ haploinsufficiency phenotype , and this resembled Notch signaling defects . Notch is required at multiple steps in the process specifying SOPs ( sensory organ precursors ) and the different cell types originating from them , including the positioning and spacing of SOPs and asymmetric activation of Notch during their multiple asymmetric divisions . Based on at what time point in this process Notch signaling is disrupted , a variety of phenotypes can be expected [52–58] . When analyzed in pupal thorax clones , relative to neighboring wild type ( WT ) , AKAP200 mutant tissue displayed loss of SOPs and SOP mispositioning ( S1E and S1F Fig ) , as well as rare SOP lineage defects ( Fig 1E and S1F and S1F’ Fig , arrowheads ) . In the wing , AKAP200 mutant wings appeared blistered as did mutant clones , but we did not observe defects to the margin ( S1J Fig ) . Interestingly , the blistered wing phenotype suggested a PKA related function of AKAP200 [59 , 60] , whereas the eye and thorax phenotypes resembled a subset of Notch LOF defects with no obvious link to PKA signaling . Taken together , these phenotypic defects suggest that AKAP200 might affect Notch signaling in the eye and thorax , whereas it seems to act ‘canonically’ via PKA in wings . The AKAP200 mutant phenotypes resembled that of Notch LOF in eyes and the thorax , suggesting a novel function for AKAP200 . To gain further insight into its potential involvement in the Notch pathway , we tested for genetic interactions between AKAP200 and Notch-associated genotypes ( LOF and GOF ) in the eye and thorax . The Drosophila eye is a compound eye with a mirror symmetric organization of ommatidia across the dorso-ventral midline [61] ( Fig 2A and 2B ) . Perturbation of Notch signaling during ommatidial assembly can lead to a variety of phenotypes [28 , 31 , 33 , 50] . These can be classified into ommatidia with the WT complement of photoreceptors , PRs ( 6 outer photoreceptors and single R7/R8 ) and include altered ommatidial orientation , flips , or clusters with a symmetrical appearance; and ommatidia with PR number defects ( for example loss or gain of R7; Fig 2A ) . To probe the relationship of AKAP200 and the Notch pathway , we asked whether AKAP200 mutants influenced Notch signaling in the eye . NΔECD , a membrane-tethered deletion of the extracellular domain that renders Notch active in a ligand-independent manner [62] , was expressed under the control of the sevenless ( sev ) promoter [which is initially expressed in R3-R4 , and later in R1 , 6 and 7 and cone cells [63]] . sev-NΔECD causes the formation of R4 symmetrical clusters and chirality flips besides frequent defects in PR number and supernumerary R7s ( Fig 2C and 2H ) . Reducing the copy number of AKAP200 in the sev-NΔECD background markedly reduced these defects . Whereas PR cell loss was reduced , a proportional increase was seen in less severe Notch GOF phenotypes , including R4 symmetrical clusters and chirality flips ( Fig 2D and 2H ) . The suppression of Notch overactivation by AKAP200M30 was comparable to reducing Notch protein levels , e . g . Notch-/+ ( Fig 2H ) . These interactions were reproducible with all AKAP200 alleles and deficiencies tested , and was strongest upon homozygous removal of AKAP200 ( AKAP200M30/M24 ) in rare escapers ( Fig 2H ) , altogether suggesting that AKAP200 promotes Notch signaling . To confirm this assessment , we tested additional phenotypes associated with the activated Notch pathway . A milder Notch GOF background , using a weaker sev-Gal4 driver ( sev-enhancer with sev promoter: “sep>” ) resulted in chirality defects with flips and R4 symmetrical clusters ( Fig 2E and 2I ) . Upon removing one copy of AKAP200 in this background , the number of R4 symmetrical clusters was markedly reduced ( Fig 2F and 2I ) , again consistent with the notion that AKAP200 promotes Notch signaling . To confirm these interactions , we next analyzed larval eye discs for the expression of Prospero ( Pros ) , a molecular marker for R7 ( and also later in cones cells ) ( Fig 2G; Elav , was used as a pan-neuronal marker to stain all PRs ) . In contrast to WT , where each ommatidium has a single Pros positive R7 , in sev-NΔECD most ommatidia had 2 or 3 Pros positive R7s . Removing a copy of AKAP200 suppressed this phenotype , with an appearance closer to WT ( Fig 2G and 2J ) . These observations are consistent with the interactions above and correlate with phenotypes seen in adult eyes ( Fig 2C , 2D and 2H ) . We next performed the equivalent experiment using a Notch signaling reporter , mδ0 . 5-lacZ , a 500 bp fragment of the E ( spl ) mδ promoter [37] , which serves as a molecular readout of Notch signaling specifically in R4 ( it is initially expressed at low levels in the R3/R4 pair , and following Notch activation it is upregulated in R4 ) . In WT , each ommatidium displayed a single mδ-lacZ positive cell , whereas , in contrast , sev-NΔECD eye discs displayed generally 2 mδ-lacZ positive cells per ommatidium . Consistent with the effects in adult eyes , removing one copy of AKAP200 in the sev-NΔECD background suppressed this phenotype , with most ommatidia displaying only one mδ-lacZ positive cell ( S2K and S2L Fig ) . In order to solidify the notion that AKAP200’s eye phenotypes are directly related to Notch signaling or whether other pathways are involved , we also tested for a potential AKAP200 involvement with Egfr signaling , which has similar phenotypes [64 , 65] . To do this , we performed genetic interactions with Egfr GOF by using the EgfrElp1/+ allele ( S2M Fig ) and saw no significant difference in its PR number defect phenotype when one copy of AKAP200 was removed . Furthermore , we performed this interaction in the wing . Again , AKAP200M30/+ ( S2O Fig ) was unable to modify the ectopic vein phenotypes of EgfrElp1/+ ( S2N Fig ) . We next wished to confirm the link between Notch and AKAP200 in the thorax , as AKAP200 LOF displayed supernumerary scutellar macrochaetae defects ( Fig 1E , red arrow ) , which highly resemble the Notch haploinsufficiency phenotype . Strikingly , an increase in Notch copy number ( 3 copies ) suppressed the macrochaetae defects of AKAP200 LOF mutants ( S3N and S3O Fig ) . In contrast , removing one copy of AKAP200 enhanced the Notch haploinsufficiency ( N55e11/+ ) thorax phenotype ( quantified in S2F Fig , depicted in S2G–S2J Fig ) . These data confirmed a positive requirement for AKAP200 in Notch signaling and suggested that AKAP200 might affect Notch levels ( see below ) . Since the N null alleles do not display haploinsufficient phenotypic defects in the eye , analogous eye experiments could not be tested . To further corroborate the link between AKAP200 and Notch signaling , we tested Notch GOF genotypes in the thorax . Overactivation of Notch signaling in the thorax can be achieved via the haploinsufficient Hairless ( H ) mutant . H is a nuclear antagonist of Notch signaling and represses Notch target genes by assembling a transcriptional repressor complex [66] . H is involved in neuronal fate specification , and the mutant thorax phenotype reflects overactivated Notch signaling during SOP specification , resulting in reduction of bristles ( quantified in S2A Fig , depicted in S2B and S2C Fig ) . Su ( H ) is the DNA-binding transcription factor that is directly bound by NICD . H antagonizes Su ( H ) ’s ability to bind to NICD and thereby activate transcription [67] , consistent with known interactions between Su ( H ) and H/+ ( quantified in S2A Fig , depicted in S2D Fig ) . Similarly , N55e11/+ suppressed H/+ albeit to a lesser degree ( S2A Fig ) . Supporting the interactions in the eye , AKAP200M30 and AKAP200M24 , as well as the AKAP200 deficiency suppressed the H/+ phenotype ( S2A Fig , depicted in S2E Fig ) . Taken together , the data from the eye and thorax are consistent with AKAP200 promoting Notch signaling activity . Since AKAP200 is known to interact with and confine the cellular localization of PKA , we next determined whether AKAP200’s effects on Notch signaling required PKA . We tested the ability to rescue the AKAP200 LOF phenotype of AKAP200-L , which binds PKA , and AKAP200-S , which does not as it lacks the PKA interaction domain ( Fig 3A ) . Strikingly , ubiquitous expression of each isoform ( under tubulin-Gal4 control; Fig 3D and 3E ) rescued the AKAP200 photoreceptor number defects ( Fig 3C ) . This suggests that eye phenotypes are not related to PKA ( quantified in Fig 3B; this also confirmed that the mutants are clean AKAP200 alleles; see also S3 Fig ) . Similarly , both isoforms were capable of rescuing the AKAP200 bristle defects ( S3K , S3M and S3O Fig ) . To lend further support to the hypothesis that AKAP200’s supernumerary bristle phenotype can be attributed to Notch signaling , we added an extra copy of Notch using N-GFP , Cherry flies [68] in an AKAP200 mutant background . Here as well we observed a rescue of the AKAP200 bristle defects ( S3N and S3O Fig ) . We also assessed the potential direct involvement of PKA in promoting Notch signaling , and asked whether sev-NΔECD ( Fig 3G ) is sensitive to PKA levels . Removing one genomic copy of PKA did not modify the sev-NΔECD phenotype ( Fig 3H ) . Moreover , simultaneous removal of one copy each of AKAP200 and PKA ( Fig 3J ) had the same effect on sev-NΔECD as removing only AKAP200 ( Fig 3I; quantified in Fig 3F ) . To lend further support to this hypothesis , we tested for potential effects of AKAP200-L and AKAP200-S on the sev-NΔECD/+ eye phenotypes ( S3C Fig ) ; sev-Gal4 driven overexpression of either AKAP200-L ( S3D Fig ) and AKAP200-S ( S3E Fig ) both enhanced sev-NΔECD to comparable extents ( quantified in S3F Fig ) ; as overexpressing AKAP200-L or S alone caused no defects in the eye ( although each of them did in the wing , S3A and S3B Fig ) , this indicated that in the eye the enhancement is not an additive effect of unrelated phenotypes . Finally , overexpression of either Notch itself ( S3H Fig ) , AKAP200-L ( S3I Fig ) or AKAP200-S ( S3J Fig ) in the entire wing blade ( nubbin-Gal4 control ) produced similar phenotypes of expanded wing veins . Taken together , these data are consistent with the notion that AKAP200’s positive role on regulation of Notch signaling is PKA independent . Next , we investigated whether AKAP200 and Notch can be present in the same protein complex . Since both AKAP200 isoforms are equivalent with respect to modulating Notch function , we performed our analyses with AKAP200-S only . A Notch encoding plasmid was transfected into S2 cells with either AKAP200-S-Flag or Flag alone . Immunoprecipitation with the Flag antibody led to the co-immunoprecipitation ( co-IP ) of the NICD fragment in the AKAP200-S-Flag sample but not the control ( Fig 4A , schematic of Notch protein in S4B Fig ) . Long exposure revealed also co-immunoprecipitation of the NEXT with AKAP200-S , but we did not detect immunoprecipitation of full length Notch ( S4A Fig ) . In an inverse experiment , an AKAP200-S-Flag encoding plasmid was transfected into S2 cells with either Notch-GFP or GFP alone . Immunoprecipitation with the GFP antibody led to the co-immunoprecipitation ( co-IP ) of AKAP200-S-Flag in the Notch-GFP sample but not the control ( Fig 4B ) . To confirm the physiological relevance of these results , we examined the localization of AKAP200-S-Flag and endogenous Notch in third instar eye and wing imaginal discs . We observed co-localization of AKAP200 and Notch ( Fig 4C , co-stained with PatJ , R = 0 . 67 , S4C Fig , co-stained with E-Cad ) . Similarly , we observed colocalization of the two proteins in the portion of the wing imaginal discs from which the thorax arises ( S4D Fig , R = 0 . 4 ) . To dissect the mechanism ( s ) underlying the genetic interactions between AKAP200 and Notch and as AKAP200 promotes Notch signaling , we tested whether AKAP200 could regulate Notch protein cleavage or levels . We did not observe any reproducible changes to the Notch cleavage patterns ( examples in Fig 5A and S5A Fig ) . However , strikingly , total endogenous Notch levels were markedly reduced in homozygous AKAP200 mutant backgrounds as compared to WT ( Fig 5A and 5C; total Notch levels are the sum of full length Notch , transmembrane Notch/NEXT , and NICD , S5A Fig ) . Conversely , overexpression of AKAP200-S caused an increase in Notch levels ( Fig 5B and 5C ) . Furthermore , RT-PCR amplification of Notch showed no significant gene expression differences in AKAP200M30 eye disc lysate relative to WT ( S5B Fig ) . Several studies have demonstrated both lysosomal and proteosomal degradation of Notch [20–24] . Thus , we postulated that AKAP200 might regulate Notch turnover . To test this hypothesis , we asked if there is differential ubiquitination of Notch in AKAP200 mutants relative to WT . We performed immunoprecipitations of Notch-Flag from eye disc lysates of WT and AKAP200M30/+ larvae . Upon immunoprecipitation with Flag ( and thus Notch ) and probing for ubiquitin , there was an increase in ubiquitinated Notch-fl , NEXT and NICD fragments in AKAP200M30/+ backgrounds ( Fig 5D ) . These observations suggest that AKAP200 promotes Notch signaling by stabilizing Notch protein levels and are consistent with the genetic interaction data and corroborated by the observation that 3 genomic copies of Notch rescue AKAP200 LOF defects . In C2C12 myoblasts , Notch has been shown previously to be targeted for lysosomal degradation as a consequence of mono-ubiquitination by the E3 ubiquitin ligase , Cbl [21] . We thus hypothesized that the reduction of Notch levels and its increased ubiquitination in AKAP200 mutants could involve Cbl . To explore a role for cbl in the interplay between AKAP200 and Notch , we first tested whether they interacted genetically . Removing one genomic copy of cbl ( Fig 6B ) did not modify the sev-NΔECD/+ phenotype ( Fig 6A ) . However , the AKAP200M30/+ suppression of sev-NΔECD/+ ( Fig 6C ) was dependent on Cbl levels , with the suppression effect being markedly reduced upon simultaneous removal of one copy of both AKAP200 and cbl ( Fig 6D , quantified in Fig 6E ) . This suggests that the positive effect of AKAP200 on Notch levels and signaling activity is through antagonizing the negative function of Cbl . To confirm this , we examined the interaction between AKAP200 , cbl and Notch in SOP specification , activating the pathway with the H1 allele and using bristle loss as the GOF assay ( quantified in S6A Fig , depicted in S6B Fig ) . Consistent with the eye data , AKAP200M30/+ suppressed the H/+ phenotype ( S6D Fig , quantified in S6A Fig ) . Simultaneously removing one copy of both AKAP200 and cbl ( S6E Fig ) dampened the effect of AKAP200’s loss on the H/+ phenotype ( S6D Fig , quantified in S6A Fig; note that cbl/+ alone as a control has no effect on H1/+ , S6C Fig ) . As previously observed ( S2F and S2I Fig ) , AKAP200M30/+ enhanced the N55e11/+ scutellar phenotype ( S6I Fig ) , and consistently with the above interactions , simultaneous removal of a genomic copy of both AKAP200 and cbl limited the effect of AKAP200M30/+ on N55e11/+ ( S6J Fig; note that N55e11/+; cbl/+ alone as a control has no effect S6G Fig ) . Taken together , these data are consistent with the notion that AKAP200 antagonizes Cbl to promote Notch stability and hence promote signaling . The Cbl docking consensus site has been mapped to the vicinity of Notch’s PEST domain ( S4B Fig ) . We thus expressed a Notch isoform truncated at amino acid 2155 [69] and thereby deleting the PEST domain ( under sev-Gal4 control ) either in WT or in the AKAP200M30 background . Unlike with full-length Notch , we detected no significant difference in the phenotypic effect between the two genotypes ( S6K and S6L Fig , quantified in S6M Fig ) . This suggests that the role of AKAP200 in promoting Notch function depends on the presence of the PEST domain , consistent with the hypothesis that AKAP200 protects Notch from Cbl-mediated ubiquitination . We next assessed the effects of AKAP200 on Cbl-mediated Notch level reduction . We compared Notch protein levels from WT eye discs to AKAP200M30/+ and AKAP200M30/+; cbl/+ discs . While reducing the copy number of AKAP200 resulted in a decrease of Notch levels , concurrent reduction in copy number of both AKAP200 and cbl largely abolished this effect ( Fig 6F , quantified in Fig 6G ) . In line with this observation , Cbl/+ suppresses the AKAP200M30 PR number defect phenotype ( S6N Fig ) . To confirm specificity of the AKAP200 effect on Cbl , we tested other E3-ubiquitin ligases for an interaction with AKAP200 . The Drosophila homolog of Sel10/Fbw7 E3-ligase , archipelago ( ago ) , is an E3 ligase that also has been shown to ubiquitinate NICD and target it for proteosomal degradation [19] . Since ago is also a transcriptional target of Notch [70] , the effect of removing one copy of ago likely affects feedback loops , and thus was not included in our analyses . Instead , we tested if the AKAP200 effect on Notch can be altered if a copy of ago is also removed together with AKAP200 . Simultaneous removal of one copy of both , ago and AKAP200 , affects the sev-NΔECD/+ phenotype to the same extent as removing only a copy of AKAP200 , implying that AKAP200 and ago act via unrelated mechanisms ( S6M Fig ) . This suggests that AKAP200’s effect is specific to Cbl’s ubiquitination of Notch . Since AKAP200 appeared to protect Notch against the effects of Cbl , which targets Notch to the lysosome , we wanted to investigate the requirement of the lysosome in AKAP200’s action on Notch . We thus analyzed the effect of AKAP200 on sev-NΔECD in the presence of the lysosomal inhibitor chloroquine , which is a lysosomotropic agent acting by increasing lysosomal pH thus inhibiting lysosomal hydrolases as well as fusion of endosomes and lysosomes and thereby impairing degradation [71–74] . The effect of AKAP200M30/+ on sev-NΔECD/+ under control conditions ( Fig 7A and 7B , quantified in Fig 7E ) was lost when larvae were raised on 1mg/ml chloroquine ( Fig 7C and 7D , quantified in Fig 7E ) . At this dosage , chloroquine by itself had negligible effects on normal development and cell viability , and PR number ( quantified in Fig 7E , S7F Fig ) . Also , the average lifespan of both sev-NΔECD/+ and sev-NΔECD/+ , AKAP200M30/+ were comparable at increasing exposures to chloroquine; survival of both genotypes dropped only at chloroquine concentrations significantly greater than 1mg/ml ( S7F Fig; confirming a specific effect at 1mg/ml chloroquine and not a general interference ) . Immunofluorescence analyses in the larval eye disc showed minimal , if any , colocalization of AKAP200 and the lysosome ( S7G Fig ) . This is consistent with the notion that AKAP200 acts on Notch to prevent its targeting to the lysosome by antagonizing Cbl-mediated ubiquitination of Notch which occurs before Notch is targeted to the lysosome . Importantly , reduction of Notch protein levels in the heterozygous AKAP200 mutant background , relative to WT , was lost upon chloroquine treatment ( Fig 7G , quantified in Fig 7F ) . This highlights AKAP200’s dependence on the lysosome to promote Notch signaling and is consistent with its antagonistic effect on Cbl-mediated ubiquitination of Notch . To confirm these results , we conducted analogous experiments in the context of SOP specification . AKAP200/+ suppression of the H/+ bristle phenotype observed under control conditions ( quantified in S7A Fig , depicted in S7B and S7C Fig ) , was largely lost upon chloroquine treatment ( S7D and S7E Fig , quantified in S7A Fig ) . Taken together , our data indicate that the mechanism by which AKAP200 promotes Notch signaling , is by protecting Notch protein from the action of Cbl , which targets Notch for lysosomal degradation .
AKAP200 was identified in a PCP-signaling mediated screen performed in the Drosophila eye [39] . However , AKAP200 LOF phenotypes resemble Notch LOF . As PCP is instructive to Notch signaling in the R3/R4 specification context in the Drosophila eye , the identification of novel Notch pathway regulators was expected and in line with previous experiments; for example , the Notch ligand Dl was identified in the screen as well [39] . Furthermore , AKAP200’s strong and specific interaction with sev-NΔECD , which is a membrane tethered , ligand-independent activated Notch , indicates that it acts on Notch itself . Due to AKAP200’s ‘canonical’ role in PKA regulation , we tested how AKAP200 acts in Notch signaling . Analyses of the two isoforms of AKAP200 , which differ in their ability to bind PKA , revealed that AKAP200’s Notch associated function is PKA independent , which was corroborated by functional rescue assays with both isoforms rescuing the AKAP200 eye phenotypes indistinguishably . The AKAP200 LOF mutants display other defects , which in some tissues are PKA associated phenotypes: for example AKAP200 mutant wings have a penetrant wing blistering phenotype , which has been observed upon disruption of the PKA pathway [59 , 60] and , similarly , AKAP200 mutant ovaries have developmental defects , which have been linked to PKA signaling [75] . Our work identifies AKAP200 as a regulator of Notch protein levels ( also below ) . However , it does not affect Notch in all tissues , and even in tissues where it is required , it is specific to a subset of Notch signaling contexts . In the eye for example , it affects Notch signaling during photoreceptor specification but not during lateral inhibition in the furrow . Strikingly , there are no effects of AKAP200 on Notch signaling mediated wing margin development , which is even a haplo-insufficient Notch phenotype [76 , 77] . Likely , the Notch signaling feedback loops at the wing margin , which also includes Wingless ( Wg ) expression [78–80] , are not sensitive to AKAP200 mediated input . However , overexpression of AKAP200 in the wing led to expansion of wing veins , a phenotype linked to Notch GOF [81 , 82] , and thus consistent with the Notch GOF effects in photoreceptor specification in the eye . Taken together , our data suggest that AKAP200 affects Notch levels in a tissue and context specific manner , rather than being a general Notch protein level regulator . How do the AKAP200 Notch signaling requirements relate to each other ? In the eye , AKAP200 LOF defects correlate with photoreceptor specification , particularly with R7 and R4 induction , and associated steps . Specification of R7 and R4 both require Notch signaling activation from neighboring R-cells , R1/6 and R8 induce R7 and R3 activates the pathway in R4 , and interestingly in both contexts Egfr/RTK signaling is also required for the specific R-cell fate [30 , 33 , 36–38 , 83] . We did not detect an interaction between AKAP200 and the GOF EgfrEllipse alleles , however , suggesting that AKAP200 does not act via Egfr . Can this correlation be related to other AKAP200 requirement contexts ? In the wing , although there is no AKAP200 effect on the margin , both overexpression and LOFs of AKAP200 affect wing vein development . Establishment of wing veins is a multi-step , multi-pathway process , involving coordination of Notch signaling and other pathways , which also include Egfr/RTK signaling [84–88] . Notch signaling causes restriction of cell fate and width of the vein [4 , 66 , 89–92] . Consistent with the defects in the eye , LOF or GOF of AKAP200 correlates with vein development or vein widening , respectively . In the thorax , Notch is required at different stages of SOP specification and AKAP200 LOF phenotypes resemble several of these . Can this be linked to Egfr signaling as well ? Previous reports have shown an involvement of Egfr/RTK signaling in promoting bristle development , where Egfr hypomorphs developed fewer bristles [93 , 94] . The Egfr requirement has been attributed to SOPs requiring Egfr-signaling to maintain wild-type levels of ac-sc expression [95] . In summary , it appears that AKAP200 affects Notch activity/levels in specific contexts and that these involve Egfr signaling in some capacity . The role of AKAP200 appears to be in stabilization of the Notch protein: there is a decrease of endogenous Notch in the AKAP200 LOF vs . an increase of Notch in AKAP200 GOF backgrounds . We also observed increased ubiquitination of Notch in AKAP200 null mutant backgrounds . Ubiquitination and subsequent degradation of cellular proteins serves as a key mechanism to regulate their activity and disruption in this process often lead to overactivation of signaling . Both AKAP200 and Notch have previously been associated with the E3 ubiquitin ligase , Cbl [21 , 96] , and strikingly Cbl has also been linked to the regulation of Egfr [97 , 98] . Thus , we explored a role for Cbl in AKAP200’s regulation of Notch . Strikingly , suppression of Notch hyperactivation by AKAP200 depends on the presence of wild-type levels of Cbl . Our studies thus suggest that AKAP200’s function is to antagonize Cbl effects on Notch ubiquitination and protein levels . Of note , it is also possible that AKAP200 modulates Notch or other components of the signaling pathway via other mechanisms . One hypothesis that leads from our work is that AKAP200 could maintain the balance between Cbl’s effects on Notch and Egfr . Since AKAP200 is a scaffolding protein , it may affect both pathways , and only processes that require balanced effects of Notch and Egfr signaling may be impacted . AKAP200 was previously identified as a positive regulator of Ras signaling [99] . However , we did not pursue AKAP200’s role in this context as we did not detect interactions with the EgfrElp allele . In conclusion , we postulate a novel mechanism of regulation of Notch signaling by AKAP200 antagonizing Cbl-mediated lysosomal degradation of Notch . This study advances our understanding of the tight regulation of Notch protein levels , which is fundamental to numerous key developmental processes and diseases .
Flies were raised on standard medium and maintained at 25°C unless otherwise indicated . All WT experiments were performed in w1118 backgrounds . The following stocks lines were used and their sources are indicated: The Gal4/UAS system [103] was employed for gene expression studies and the following Gal4 drivers were used: sep-Gal4 [104] , sev-Gal4 [63] , tub-Gal4 , nub-Gal4 ( Bloomington stock center ) . sev-gal4 ( sev-enhancer with heat-shock promoter ) initially drives expression in the R3/R4 precursor and later in R1/6 and R7 ( note , there is basal expression in other tissues due to the presence of the heat shock promoter from hsp70 ) . sep-gal4 which has the sev-enhancer with sev promoter , results in weaker expression levels than sev-gal4 . AKAP200M30 clones were produced by mitotic recombination via the FLP/FRT system [47] with eyFLP in an AKAP200M30 FRT40A/ w armlacz FRT40A background and ubxFLP in AKAP200M30FRT40A/ y FRT40A background . AKAP200M30 and AKAP200M24 were generated using a FLP-recombinase-mediated excision of two piggybac/FRT insertions grkf07069 and AKAP200d03938 and characterized by PCR as previously described [46] . To generate UAS-AKAP200-Flag transgenic flies , the Flag tag was added to the C-term of AKAP200 sequence by PCR amplification using DGRC LD42903 and RE01501 cDNA clones for AKAP200-L and AKAP200-S respectively . The PCR amplified products were cloned into pUASt-attB vector using EcoRI and XhoI sites . The following primers were used: 5’-CCGGAATTCATGGGTAAAGCTCAGAGCAA-3’and 5-CCGCTCGAGCTTGTCGTCGTCGTCCTTGTA-3’ Transgenic injections were performed by BestGene Inc . where the constructs were targeted to predetermined genomic sites on chromosome 3R using the phiC31 integrase ( strain 9744 ) . For drug treatments , crosses were setup on instant food ( Carolina Biological Supply Company ) to which chloroquine diphosphate salt ( Sigma ) or water was added at the indicated concentrations . To compare relative amount of Notch mRNA , RNA was extracted from eye disc lysates from WT or AKAP200M30 flies using RNeasy Mini Kit as per manufacturer’s protocol ( Qiagen ) . 1 ng of RNA was reverse transcribed ( 50°- 30’ and 94°- 2’ ) and real-time PCR was performed using SYBR Green I Master ( Roche ) on LightCycler 480 ( Roche ) . Quantification was performed using the 2-ΔΔCT method and Gapdh transcript as a reference . Measurements were performed in duplicate . The following primers were used: The amplified products are expected to be ~300 bp for rp49 and ~1600 bp for Notch . As control for DNA contamination in eye disc lysates , a reaction was run using Notch primers excluding reverse transcriptase ( rxn mixture ) . Third larval instar eye discs were dissected in ice cold PBS and fixed in PBS-4% paraformaldehyde for 20 minutes at room temperature . After three washes in PBT ( PBS + 0 . 1% Triton-X ) , discs were incubated in primary antibody overnight at 4°C . After three PBT washes , secondary antibody incubation for 2 hours at room temperature and three more PBT washes , the discs were mounted in Vectashield ( Vector Laboratories ) . For immunofluorescence , the following antibodies were used- mouse anti-Prospero ( #MR1A , 1:10 , Developmental Studies Hybridoma Bank-DSHB ) , rat anti-Elav ( #9F8A9 , 1:20 , DSHB ) , mouse anti-NotchICD ( #C17 . 9C6 , 1:10 , DSHB ) , rabbit anti-Flag ( #637301 , 1:100 , Biolegend ) , mouse anti-Flag ( #F1804 , 1:1000 , Sigma ) , rat anti-DE Cadherin ( #5D3 , 1:20 , DSHB ) , rabbit anti-Patj ( 1:500 ) , rabbit anti-GFP ( #1828014 , 1:1000 , invitrogen ) , rabbit anti-β-gal ( 1:1000 , Molecular probes ) . Fluorescent secondary antibodies came from Jackson Laboratories . Eye disc and thorax images were acquired at room temperature using a Zeiss LSM 880 or Leica SP5 DMI confocal microscopes Subsequent image processing was performed on ImageJ ( National Institute of Health ) . For colocalization analyses , the JaCoP plugin was used in ImageJ to calculate the Pearson’s coefficient ( R ) . Eye sections were prepared as previously described [105] . All eyes were sectioned near the equatorial region . For analysis of adult thoraces , whole flies were incubated in 70% ethanol and mounted on gelatin plates . Imaging was done using a stereomicroscope and acquired using Zeiss Axioplan color type 412–312 ( Carl Zeiss ) camera and Zen Blue software . For analysis of adult wings , wings were removed and incubated in PBT , mounted on a slide in 80% glycerol and imaged using a Zeiss Axioplan microscope ( Carl Zeiss ) . For analysis of Notch protein levels , 10–15 pairs of larval eye discs were lysed in ice cold lysis buffer ( 50mM Tris HCl pH 7 . 5 , 150mM NaCl , 1mM EDTA and 1% Triton-X ) . Supernatant from these extracts were resolved and subjected to standard western blotting procedures using mouse anti-NotchICD ( #C17 . 9C6 , 1:500 , DSHB ) and mouse anti-Y-Tubulin ( Sigma Aldrich , 1:1000 ) antibodies . For co-immunoprecipitation assays , S2 cells were transfected with pmt-Notchfull length ( #1022 from DGRC ) with pAC-Flag control or pac-AKAP200-S-Flag , or pac-AKAP-200-S-Flag with pUAST-GFP or pUAST-Notch-GFP ( gift from Dr . Shigeo Hayashi ) , both with Actin-Gal4 . Transfections were performed using Effectence ( QIAGEN , Hilden , Germany ) in accordance with manufacturer’s instructions . For pmt-Notch , Notch was induced using 600 μM CuSO4 24 hours after transfection for 24 hours [100] . For Flag IPs and ubiquitin assays , cells were harvested , washed and lysed in ice-cold lysis buffer ( 50mM Tris HCl pH 7 . 5 , 150mM NaCl , 1mM EDTA and 1% Triton-X ) . For ubiquitin assays , the lysis buffer was supplemented with 100 μg/mL leupeptin ( Roche 1017101 ) and protease inhibitor cocktail tablets ( Roche , 14268500 ) . For Flag IPs , lysed samples were incubated overnight at 4°C using anti-Flag M2 agarose beads ( Sigma , A2220 ) . Beads were washed three times and protein was eluted by boiling in Laemmli buffer . For GFP IPs , cells were harvested , washed and lysed in ice-cold lysis buffer ( 0mM TrisHCl pH7 . 5 , 150mM NaCl , 0 . 5mM EDTA , 1% TritonX ) . Following this , we used the GFP-Trap A from Chromotek as per manufacturers protocol ( wash buffer: 10mM TrisHCl pH7 . 5 , 150mM NaCl , 0 . 5mM EDTA ) . Western blots were carried out with the immunoprecipitated samples using mouse anti-NotchICD ( #C17 . 9C6 , 1:500 , DSHB ) , mouse anti-Flag ( #F1804 , 1:1000 , Sigma ) , mouse anti-Ubiquitin ( #MA1-10035 , 1:5000 , Thermofisher ) and rabbit anti-GFP ( #1828014 , 1:1000 , Invitrogen ) . HRP coupled secondary antibodies were obtained from Jackson laboratories . | AKAP200 belongs to a family of scaffolding proteins best known for their regulation of PKA localization . In this study , we have identified a novel role of AKAP200 in Notch protein stability and signaling . In Drosophila melanogaster , AKAP200’s loss and gain-of-function ( LOF/GOF ) phenotypes are characteristic of Notch signaling defects . Furthermore , we demonstrated genetic interactions between AKAP200 and Notch . Consistent with this , AKAP200 stabilizes the endogenous Notch protein and limits its ubiquitination . AKAP200 exerts its effects on Notch by antagonizing Cbl-mediated ubiquitination and thus lysosome targeting of Notch . Based on these data , we postulate a novel PKA independent mechanism of AKAP200 to achieve optimal Notch protein levels , with AKAP200 preventing Cbl-mediated lysosomal degradation of Notch . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"lysosomes",
"animals",
"notch",
"signaling",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"immunoprecipitation",
"experimental",
"organism",
"systems",
"co-immunoprecipitation",
"eyes",
"g... | 2018 | AKAP200 promotes Notch stability by protecting it from Cbl/lysosome-mediated degradation in Drosophila melanogaster |
In plants , multiple detached tissues are capable of forming a pluripotent cell mass , termed callus , when cultured on media containing appropriate plant hormones . Recent studies demonstrated that callus resembles the root-tip meristem , even if it is derived from aerial organs . This finding improves our understanding of the regeneration process of plant cells; however , the molecular mechanism that guides cells of different tissue types to form a callus still remains elusive . Here , we show that genome-wide reprogramming of histone H3 lysine 27 trimethylation ( H3K27me3 ) is a critical step in the leaf-to-callus transition . The Polycomb Repressive Complex 2 ( PRC2 ) is known to function in establishing H3K27me3 . By analyzing callus formation of mutants corresponding to different histone modification pathways , we found that leaf blades and/or cotyledons of the PRC2 mutants curly leaf swinger ( clf swn ) and embryonic flower2 ( emf2 ) were defective in callus formation . We identified the H3K27me3-covered loci in leaves and calli by a ChIP–chip assay , and we found that in the callus H3K27me3 levels decreased first at certain auxin-pathway genes . The levels were then increased at specific leaf genes but decreased at a number of root-regulatory genes . Changes in H3K27me3 levels were negatively correlated with expression levels of the corresponding genes . One possible role of PRC2-mediated H3K27me3 in the leaf-to-callus transition might relate to elimination of leaf features by silencing leaf-regulatory genes , as most leaf-preferentially expressed regulatory genes could not be silenced in the leaf explants of clf swn . In contrast to the leaf explants , the root explants of both clf swn and emf2 formed calli normally , possibly because the root-to-callus transition bypasses the leaf gene silencing process . Furthermore , our data show that PRC2-mediated H3K27me3 and H3K27 demethylation act in parallel in the reprogramming of H3K27me3 during the leaf-to-callus transition , suggesting a general mechanism for cell fate transition in plants .
Unlike most animal tissues , a wide variety of plant tissues can easily acquire pluripotency when properly cultured in vitro with two plant hormones , auxin and cytokinin [1] . On a callus-inducing medium ( CIM ) , explants usually first form a pluripotent cell mass , called callus , from which shoots and roots regenerate on the corresponding shoot- or root-inducing medium [2] . Recent studies revealed that callus formation is via a lateral root development pathway involving a process where cells evolve from pericycle-like to form root meristem-like cells [3]–[5] . This novel finding greatly improves our understanding of the plant cell regeneration process , but also raises a new question: what is the underlying mechanism that guides cell fate transition from diverse differentiated plant tissues to callus ? Callus formation from explants requires dramatic changes both in cell identities and cell growth patterns , and such a cell fate transition has been shown to be accompanied by changes in expression of numerous genes [3] , [5]–[7] . It seems unlikely that the genome-wide changes in gene expression only involve spatially and temporally regulated transcription factors . Plant epigenetic pathways , which are known to influence genome-wide gene expression [8] , may also participate in the large-scale gene regulations occurring during callus formation [7] , [9] . Chromatin , which is composed of repeating units termed nucleosomes , is the template of epigenetic information , and changes in chromatin structure could lead to simultaneous expression changes of numerous genes [8] , [10] . Changes of histone modifications often affect epigenetic regulation [11] . In Arabidopsis thaliana , silenced euchromatin is usually marked with histone H3 lysine 27 trimethylation ( H3K27me3 ) [12]–[14] , while the active gene transcription phase is usually associated with H3K4me3 and H3K36me3 [14]–[16] . We show in this work that reprogramming of H3K27me3 is critical in genome-wide regulation of key genes , leading to cell fate transition from leaf blade to callus . Reprogramming of H3K27me3 involves the Polycomb group ( PcG ) -mediated H3K27 trimethylation and H3K27 demethylation . PcG proteins form at least two multiprotein complexes , namely Polycomb Repressive Complex 1 ( PRC1 ) and PRC2 . While PRC2 shows histone methyltransferase activity , targeting lysine 27 on the N-tail of histone H3 , PRC1 recognizes the trimethylation marker for subsequent chromatin compression [17] , [18] . The Arabidopsis proteins CURLY LEAF ( CLF ) , SWINGER ( SWN ) and MEDEA [19]–[21] were proposed to be the core components of PRC2 , and the H3K27 methyltransferase activity of CLF was shown in a biochemical assay [22] . In addition to the core components , PRC2 also contains other components , including EMBRYONIC FLOWER2 ( EMF2 ) in Arabidopsis [21] , [23] . On the other hand , RELATIVE OF EARLY FLOWERING 6 ( REF6 ) is an H3K27me3 demethylase identified in Arabidopsis , which efficiently removes the methyl group from H3K27me3 both in vitro and in planta [24] . However , since the ref6 mutation results in H3K27me3 hypermethylation only on a part of the PRC2-targeted loci [24] , it is possible that other unidentified H3K27me3 demethylase ( s ) or additional demethylation mechanism ( s ) exist . In this study , we show that genome-wide reprogramming of H3K27me3 is critically required for the leaf-to-callus transition . The PcG pathway is responsible for repression of the leaf-regulatory genes in leaf blade explants , and acts in parallel with the Arabidopsis H3K27 demethylation pathway , which derepresses the auxin-pathway and root-regulatory genes to enable the leaf-to-callus transition .
Because the gene expression profiles in calli differ dramatically from those in their original tissues [3] , [5]–[7] , we hypothesized that one or more epigenetic pathways , which function in genome-wide regulation of gene expression , may participate in this cell fate switching process . To test this hypothesis , we first analyzed callus formation using leaf blades from mutants with reduced levels of H3K4me3 , H3K36me3 , or H3K27me3 , which in Arabidopsis are important for gene regulation in the euchromatin regions [11] , [14] . For convenience , explants of leaf blade are referred to as leaf explants hereafter . The mutants used corresponded to the methyltransferases of ARABIDOPSIS THALIANA TRITHORAX1 ( ATX1 ) and SET DOMAIN GROUP2 ( SDG2 ) for H3K4me3 [25]–[27] , SDG8 for H3K36me3 [28] , [29] , and CLF and SWN for H3K27me3 [12] . Compared with the wild type ( Figure 1A , 1E ) , atx1-2 , sdg2-3 , and sdg8-2 leaf explants formed calli on CIM ( Figure 1B–1D ) , whereas no callus was seen from leaf explants of the clf-50 swn-1 double mutant ( Figure 1F ) . It should be mentioned that swn-1 is a weak allele , and the null clf swn double mutant displays distinct plant phenotypes [21] . Because both CLF and SWN are core components of PRC2 and are functionally redundant , these results suggest that the PRC2-mediated H3K27me3 is required for the leaf-to-callus transition . Since H3K27me3 is required for callus formation from leaves , we next analyzed the genome-wide H3K27me3 profile by a ChIP-chip assay to identify genes with the altered H3K27me3 modification between the leaf and callus ( Figure 2A; Table S1 ) . A total of 3856 and 3991 genes harboring high levels of H3K27me3 were identified in the leaf and callus , respectively , with many of these included among the 4979 genes previously identified in seedlings [13] ( 72 . 7% and 75 . 6% , respectively ) ( Figure 2B ) . Additionally , 2306 genes were common among the leaves , calli , and seedlings ( Figure 2B ) . Furthermore , 434 and 186 genes showed considerable and significant decrease and increase in H3K27me3 levels , respectively , in the callus as compared to the leaf ( Figure 2C , Table S1 ) . Genes bearing the altered H3K27me3 levels encode proteins with different functions ( Figure 2C ) , including phytohormone pathway proteins and putative transcription factors . It is possible that some of these putative regulatory genes may have important roles during callus formation , and we thus analyzed expression profiles of these putative regulatory genes in four different tissues: seedlings , inflorescences , rosette leaves , and roots , based on the data from published databases [30] . Compared with leaves , calli contained the increased number of genes which are normally expressed in the root of a plant , and all these genes belonged to the category with decreased H3K27me3 levels ( Figure 2D , left ) . Conversely , among the H3K27me3 level increased category , the number of genes normally expressed in the root of a plant was reduced , whereas the number of genes expressed in the leaf was increased ( Figure 2D , right ) . Since the H3K27me3 modification is generally thought to negatively affect gene expression [12] , [13] , these results support the previous notion that the callus cells are the root-featured cells [5] . We also performed a microarray experiment to analyze whether and to what degree the changed H3K27me3 levels of these genes affect their expression levels . Our data showed that about 40% genes with decreased H3K27me3 levels were upregulated , and about 46% genes with increased levels of H3K27me3 were downregulated ( Table S2 ) . This result indicates that the H3K27me3 modification and gene expression regulation are correlated for a relatively large group of the H3K27me3-covered genes during the leaf-to-callus transition . To better understand callus development , we analyzed phenotypes of leaf explants during culturing using scanning electron microscopy ( SEM ) . Shapes of leaf explants on CIM were unchanged 2 days after culturing ( DAC ) ( Figure 3A ) , compared with explants prior to culturing ( i . e . , time 0 ) ( Figure S1 ) . Additionally , cell proliferation was not observed in the margin of 2 DAC explants ( Figure 3A , 3B ) . In 4 DAC explants , cells started to proliferate , initially in the middle of the midvein ( Figure 3C , 3D ) . Six days after culturing , cell proliferation occurred in the margin of leaf explants and vigorous cell division surrounding the midvein in the proximal part of the explants was observed ( Figure 3E , 3F ) . Cell proliferation was clearly accelerated in 8 DAC and 10 DAC explants ( Figure 3G , 3H ) , and the newly formed calli from the midvein and margins covered most parts of the 10 DAC explants ( Figure 3H ) . The CYCB1;1 gene is strictly regulated by the cell cycle , and thus is widely used as a cell division marker [31] . By analyzing a CYCB1;1:GUS transgenic line , we demonstrated that the CYCB1;1 expression pattern was consistent with the results from SEM . The 2 DAC leaf explants did not show any CYCB1;1:GUS staining ( Figure 3I ) , whereas GUS staining was observed in the midvein and other fine veins in the proximal part of 4 DAC explants ( Figure 3J ) . The strong CYCB1;1:GUS staining was observed along the midvein and vein branches , and in the margins of the 6 DAC and 8 DAC explants ( Figure 3K , 3L ) . Based on the time of cell proliferation initiation , callus formation from leaf explants could be mainly divided into two stages: the pre- and post-2 DAC stages , reflecting prearrangement and performance of cell fate transition , respectively . Auxin is a plant hormone required for callus induction [2] , and our ChIP-chip data revealed that H3K27me3 levels of a group of auxin-pathway genes were dramatically reduced in the callus compared with the leaf . These included several auxin-inducible genes , such as the GH3 genes , GH3 . 1 , 2 , 3 , 6 , and 17 , which are involved in auxin homeostatic control , and the AUXIN/INDOLE-3-ACETIC ACID ( AUX/IAA ) genes , IAA1 , 2 , 14 , 19 , 20 , and 24 ( Table S1 ) , which participate in auxin signaling [32] , [33] . Our qRT-PCR results revealed that these genes were induced before or on 2 DAC . For example , compared with the basal expression levels in time 0 explants , GH3 . 2 expression levels were dramatically increased in 2 DAC explants and showed the highest levels in the 4 DAC explants ( Figure 4A ) . The highest level of IAA2 expression also appeared on 2 DAC ( Figure 4A ) . We also analyzed the time course of the H3K27me3 level changes at the GH3 . 2 and IAA2 loci by ChIP ( Figure 4B ) , and results were normalized against those at AGAMOUS ( AG ) because the H3K27me3 levels at the AG locus kept unchanged between the leaf and callus ( Figure S2 ) . Our results showed that increased expression levels of these auxin-pathway genes were accompanied by a sharp decrease in H3K27me3 levels at their loci starting from 2 DAC ( Figure 4B ) . This suggested a dramatically altered auxin-related action in the pre-2 DAC explants . The expression patterns of GH3 . 2 and IAA2 were consistent with those of the DR5pro:GUS reporter line [34] , in which GUS staining was hardly detected at time 0 ( Figure 4C ) , but was evident in the 2 DAC explants ( Figure 4D ) . The strongest GUS staining was observed in 4 DAC explants ( Figure 4E ) . Compared with the time 0 explants , which only showed very faint GUS signals surrounding the midvein ( Figure 4C ) , deep GUS staining of the 2 DAC DR5pro:GUS explants was detected in the margins and midveins ( Figure 4D ) . As cells were rapidly proliferating in 6 DAC and 8 DAC explants , all freshly formed calli were deeply stained ( Figure 4F , 4G ) , indicating that a strong auxin signal transduction was occurring in these tissues . In addition to the GH3 . 2 and IAA2 genes , other genes related to auxin biosynthesis , metabolism , and transport showed a significant loss in H3K27me3 ( Figure 4H , Table S1 ) . These genes included YUCCA4 ( YUC4 ) [35] , NITRILASE2 ( NIT2 ) [36] , IAA CARBOXYL METHYLTRANSFERASE1 ( IAMT1 ) [37] , and PIN-FORMED1 ( PIN1 ) [38] . These results suggest that the entire network of auxin regulation is activated during the leaf-to-callus process . To understand the molecular basis during the leaf-to-callus transition , we analyzed expression patterns of all putative transcription factor genes that showed altered H3K27me3 levels in the ChIP-chip assay , as these genes may include important contributors to cell fate transition . The analysis was first performed using Genevestigator , which is a program available online ( https://www . genevestigator . com/ ) [30] . Interestingly , many of these transcription factor genes with decreased H3K27me3 levels were those that were either silenced or exhibited low-level expression in the leaf , but were highly expressed in the roots ( Figure 5A ) . Conversely , most genes that showed increased H3K27me3 levels were those that were originally predominantly expressed in the leaf , but were silenced or weakly expressed in the roots ( Figure 5A ) . These results indicate that callus formation from leaf explants experiences a tissue identity transition from leaf to root , and thus support the recent proposal that the callus contains cells resembling the root meristematic pluripotent cells [5] . Our data also suggest that reprogramming of H3K27me3 is important for changes in expression of regulatory genes that are originally preferentially expressed in specific tissues . We also analyzed gene expression levels over time by qRT-PCR and monitored H3K27me3 levels by ChIP for several selected marker genes of transcription factors during callus formation . SAWTOOTH1 ( SAW1 ) and SAW2 are genes that control leaf development [39] and were selected as the leaf marker genes . Unlike the analyzed auxin-pathway genes , expression levels of SAW1 and SAW2 were largely unchanged 2 DAC , but were reduced markedly 4 DAC and continued to gradually decrease during the subsequent post-2 DAC stages ( Figure 5B ) . The reduced SAW1 and SAW2 expression levels were accompanied by increased H3K27me3 levels beginning 4 DAC ( Figure 5C , 5D ) . The increased H3K27me3 levels were observed for some other leaf development genes , such as the BELL family gene ARABIDOPSIS THALIANA HOMEOBOX GENE1 ( ATH1 ) [40] and the TEOSINTE BRANCHED1-CYCLOIDEA-PCF ( TCP ) transcription factor gene TCP10 [41] , [42] ( Figure 5D , Table S1 ) . These results suggest that PRC2-mediated H3K27me3 plays a role in silencing leaf regulatory genes during callus formation . WUSCHEL-RELATED HOMEOBOX 5 ( WOX5 ) is a root gene specifically expressed in the quiescent center ( QC ) and SHORT-ROOT ( SHR ) is an important root development-controlling gene [43] , [44] . WOX5 was slightly derepressed in the 4 DAC explants and was highly expressed in 6 DAC and 8 DAC explants . A drastic SHR derepression also occurred 6 DAC ( Figure 6A ) . The increased WOX5 and SHR expression levels were consistent with the decreased H3K27me3 levels starting 4 DAC ( Figure 6B ) , suggesting that H3K27 demethylation is also required for derepression of root genes . We then used WOX5pro:GUS and SHRpro:GUS transgenic plants to analyze their expression patterns during callus formation . In the WOX5pro:GUS transgenic line , GUS staining was only detected in the root QC cells ( Figure 6C ) [43] . WOX5 remained silenced in leaf explants 2 DAC ( Figure 6D ) , but expression in the midvein was observed 4 DAC ( Figure 6E ) . GUS staining in 6 DAC and 8 DAC explants was observed in the midvein and the fine vein ends close to the wounded margins in emerging calli . This staining became more intense with increasing cell proliferation ( Figure 6F , 6G ) . These results indicate that many cells in the newly formed calli from leaf explants possess features of root QC cells . SHR is also expressed in the leaf vascular system ( Figure 6I ) in addition to its expression in the root stele cells ( Figure 6H ) [44] . However , SHR expression in early-stage explants is obviously different from that in the late-stage explants . The SHR expression pattern in the leaf vein did not change from time 0 to 4 DAC ( Figure 6I–6K ) . In the 6 DAC leaf explants , SHR expression was weakened in the original leaf veins but became more intense in the midvein and vein ends near the wounded margin of newly formed calli ( Figure 6L , 6M ) . In addition to WOX5 and SHR , the H3K27me3 levels of some other root genes , such as LATERAL ORGAN BOUNDARY DOMAIN33 ( LOB33 ) [45] and AGAMOUS-LIKE21 ( AGL21 ) [46] , were also consistently decreased in the post-2 DAC stage ( Figure 6N , Table S1 ) . Changes in expression and H3K27me3 levels of the analyzed leaf and root regulatory genes highlight the process of cell fate transition occurring during callus formation in the post-2 DAC stage , following the auxin-pathway activation in the pre-2 DAC stage . To determine the role that the PRC2 plays in callus formation , we analyzed the expression of selected regulatory genes in the clf-50 swn-1 mutant . Both GH3 . 2 and IAA2 showed a similar elevated expression compared to the wild-type counterparts , suggesting that auxin response may not involve PRC2 ( Figure 7A , 7B ) . Expression levels of SAW1 and SAW2 were reduced ( Figure 7C , 7D ) and those of WOX5 and SHR were elevated ( Figure 7E , 7F ) in 2 DAC leaf explants of clf-50 swn-1 , rather than the expression changes occurring in 4 DAC leaf explants in the wild type . This was indicative of a shortened pre-2 DAC stage in the PRC2 mutant . Interestingly , expression levels of SAW1 and SAW2 in clf-50 swn-1 leaf explants failed to continue to decrease during the post-2 DAC stage ( Figure 7C , 7D ) , as was observed in the wild type ( Figure 5B ) , but instead gradually increased . On the other hand , the H3K27me3 modification at the SAW1 and SAW2 loci consistently retained at a very low level in clf-50 swn-1 ( Figure S3 ) . These results indicate that the leaf explants of clf-50 swn-1 consistently retain their leaf identities , although WOX5 and SHR were derepressed during culturing ( Figure 7E , 7F ) . Because PcG is required for silencing of the leaf development-controlling genes , but does not affect the derepression of auxin- and root-related genes , it is possible that leaf and root explants of the PRC2 mutant might behave differently during callus formation . To test this hypothesis , we analyzed the ability of the clf-50 swn-1 roots to form calli . Similar to the wild type roots ( Figure 8A ) , the clf-50 swn-1 root explants demonstrated normal callus formation ( Figure 8B ) . A more detailed analysis using SEM showed that the progression of callus formation in wild type ( Figure 8C , 8D ) and clf-50 swn-1 ( Figure 8E , 8F ) was also similar , as callus cells were evident on both root explants 6 DAC . These results indicate that PcG is required for the leaf , but not the root , during callus formation .
Recent studies have revealed that callus formation occurs through a lateral root development pathway [5] . In this study , we tried to address another layer of questions during callus formation: what is the molecular mechanism that guides different plant tissues to form the pluripotent root meristem-like cells . We show that PcG-mediated epigenetic regulation is an essential step in callus formation , and the PRC2 components CLF and SWN are required for the leaf-to-callus transition . In addition to the rosette leaves , we analyzed cotyledons of clf-50 swn-1 , and they were also defective in forming callus ( Figure S4 ) . To determine whether PRC2 as a whole is required for cell fate changes , we analyzed another PRC2 component EMF2 [21] , [23] by examining the regeneration ability of emf2 cotyledons , because of a lack of leaves in emf2 . Similar to those of clf-50 swn-1 , cotyledons from two different emf2 alleles , emf2-1 and emf2-11 , were both defective in callus formation ( Figure S4 ) , indicating the importance of PRC2 complex in the leaf-to-callus transition . Our ChIP-chip assay revealed that reprogramming of H3K27me3 occurred at several important transcription factor genes , which regulate plant meristematic functions . For example , SAW1 , SAW2 , and TCP genes are known to control the leaf margin by repression of leaf meristematic features [39] , [42] , and WOX , KNOX , and several BELL family genes are also known to be involved in root or shoot meristematic activities ( Table S1 ) [47] , [48] . It is possible that maintenance of the undifferentiated state of cells requires functions of these genes . In addition , a number of metabolism-related genes showed notable changes in the H3K27me3 level , probably reflecting a secondary effect during changes of cell fates . Interestingly , transposable elements are generally much less regulated by PRC2-mediated H3K27me3 [13] , whereas H3K27me3 levels at many transposable elements were elevated during callus formation ( Figure 2C; Table S1 ) . It is of interest to test in the future whether reprogramming of H3K27me3 at the loci of these transposable elements contributes to the callus formation . The PcG pathway plays an important role in the leaf-to-callus transition; however , since PcG regulates many target genes during this process , the exact molecular mechanism still remains elusive . The fact that many leaf genes fail to be repressed in leaf explants of the clf-50 swn-1 double mutant could be one possible reason for the defective callus formation from leaf blades , and several lines of evidence support this hypothesis . First , during callus formation from certain aerial organs of a plant , the differentiated tissues must undergo a common process to eliminate their original characteristics [5] . PcG has long been known to function in switching cell features not only in plants but also in animals . Second , most genes with increased H3K27me3 levels during callus formation are those that are highly expressed in the leaf but weakly in the root , including SAW1 , SAW2 and many others ( Figure S5 ) . Consistent with SAW1 and SAW2 , these genes demonstrated insufficient repression in leaf explants of clf-50 swn-1 ( Figure S5 ) . Third , similar to those of wild type , leaf explants of clf-50 swn-1 also show a gradually increased root gene expression , such as the demonstrated WOX5 and SHR genes , and their expression levels between wild-type and clf-50 swn-1 leaf explants are also similar within 4 DAC ( Figure 6A , 7E and 7F ) . Although after 6 DAC , expression levels of these root genes became much higher in wild type than in the double mutant , this dramatic increase of expression levels might result from the newly formed calli but not from the leaf explants , as the clf-50 swn-1 explants failed to form callus . Thus , the clf-50 swn-1 mutant has the same potential to express the root genes but is unable to silence the leaf genes to suspend the leaf program in the leaf explants . The obstacle that must be overcome for callus formation in the leaf explants does not exist in the root explants of clf-50 swn-1 and emf2 , providing a possible explanation why clf-50 swn-1 and emf2 root explants possess the ability to form calli ( Figure 8B; Figure S4 ) . Although the elimination of leaf characteristics appears to be a necessary step for callus formation from leaf explants , we could not exclude the possibility that this process is controlled by some other not-yet-revealed mechanisms . To identify key genes that preferentially expressed in the leaf and may contribute to the leaf characteristic , we overexpressed ATH1 and SAW1 and analyzed callus formation of leaf explants from 35Spro:ATH1 and 35Spro:SAW1 transgenic plants . However , overexpression of the genes did not affect the leaf-to-callus transition ( Figure S6 ) . It is possible that for maintenance of leaf characteristics , many other leaf-expressed genes may be equally important . The clf-50 swn-1 double mutant flowers earlier [21] , indicating a shortened vegetative growth phase in clf-50 swn-1 . This phenotype of clf-50 swn-1 raises a question: whether the clf-50 swn-1 leaves in ages are all equivalent to the very old wild-type ones that may lose the ability to form calli . We thus tested callus formation using wild-type and clf-50 swn-1 tissues at different ages , and found that cotyledons and leaf blades from the 44-day-old Col-0 seedlings were able to form calli , whereas the young 9-day-old leaf blades of the clf-50 swn-1 seedlings were defective in callus formation ( Figure S6 ) . These results suggest that the defective PcG function but not the age of tissues in clf-50 swn-1 blocked the callus formation . During leaf-to-callus transition , the CLF and SWN proteins may act redundantly in regulating downstream genes , as the leaf explants of both clf-50 and the stronger swn-21 single mutants normally form calli ( Figure S7 ) . The weak clf-50 swn-1 double mutant usually produces only four rosette leaves . We tested each of these leaves at different leaf developmental stages and found the phenotype of defective regeneration was very consistent ( Figure 1F; Figure S7 ) . Similar to rosette leaves , cotyledons of clf-50 swn-1 , emf2-1 and emf2-11 are also defective in callus formation ( Figure S4 ) . Although rosette leaves and cotyledons are different plant organs , they may partially share a common program , which must be terminated by PcG for regeneration . Whether the PRC2-mediated H3K27me3 is responsible for initiating or maintaining leaf gene silencing during the leaf-to-callus transition is not yet clear . It has recently been suggested that in both animals and plants , histone modifications might not initiate gene expression regulation [12] , [17] , [49] . In Arabidopsis , it was found that H3K27me3 alone is not sufficient to initiate the repression of the AG gene [12] , [17] . Our data showed that the leaf genes , SAW1 and SAW2 , were downregulated in the 2 DAC leaf explants of clf-50 swn-1 , whereas such a downregulation could not be maintained at or following 4 DAC . One possible explanation for this is that the PRC2-mediated H3K27me3 is responsible for maintaining rather than initiating the silencing of leaf genes . A leaf-to-callus transition normally requires both repression of leaf-regulatory genes and activation of auxin-pathway and root-regulatory genes . The PcG-mediated histone modification and the H3K27 demethylation pathways may act in parallel in reprogramming of H3K27me3 during callus formation . How the pre- and post-2 DAC stages are connected is not known . Whether the activated auxin pathway in the pre-2 DAC stage induces a subsequent reprogramming of H3K27me3 at the loci of the leaf- and root-regulatory genes is an interesting question that will be addressed in the future . Removal of the H3K27me3 marker may depend on an active process involving H3K27 demethyltransferases , such as REF6 [24] , or a passive process resulting in the loss of the maintenance of histone methylations during cell division [50] . We propose that in the pre-2 DAC stage , H3K27me3 demethylation of the auxin-pathway genes is mainly via an active process because during this stage , cell division does not occur . However , during the post-2 DAC stage when cells start to proliferate , H3K27 demethylation leading to derepression of the root-regulatory genes may be via both active and passive processes . We tried to address this question by analyzing leaf explants of the ref6-1 mutant , but found that normal ref6-1 calli formed from leaf explants ( Figure S7 ) . These results suggest that REF6 may not be the only demethyltransferase in callus formation or may not participate in callus formation at all , and/or passive removal of H3K27 methylation may occur . Elucidation of the molecular mechanism of H3K27me3 demethylation during callus formation is important to fully comprehend plant cell regeneration . In animals , cell fate determinations also require the PcG function for acquisition of cell pluripotency [51] , [52] . In plants , determination of flowering time , floral organ formation , and stem cell restriction are all related to reprogramming of H3K27me3 and require PcG [17] , [19] , [21] , [53]–[55] . Reprogramming of H3K27me3 might represent a general mechanism for cell fate transition in multicellular eukaryotes and the PcG pathway is at least partially involved in the process .
Seeds of clf-50 swn-1 is in the Ws background , and swn-1 is a weak swn allele [21] . Seeds of sdg2-3 [26] , sdg8-2 [28] , atx1-2 [56] and ref6-1 [57] , swn-21 ( GK-783A01 ) , emf2-11 ( GK-685A01 ) [58] , emf2-1 [59] , clf-29 [60] , and the CYCB1;1:GUS [31] and DR5pro:GUS [34] transgenic plants are in the Col-0 background . For callus induction , seeds were first grown on the medium containing Murashige and Skoog ( MS ) basal salt mixture . Tissues from different organs were prepared from different ages of seedlings and were placed on CIM ( Gamborg's B5 medium with 0 . 5 g/L MES , 2% glucose , 0 . 2 mmol/L kinetin , and 2 . 2 mmol/L 2 , 4-dichlorophenoxyacetic acid and 0 . 8% agar ) [61] , followed by incubation at 22°C in the dark . DNA fragments with 4 . 6 kb and 2 . 5 kb in length upstream to the translation initiation site of WOX5 and SHR were PCR amplified from wild-type plants , respectively , and were subcloned into the plant transformation vector pBI101 ( Clonetech , USA ) with the 3′ in-frame fusion to GUS to yield WOX5pro:GUS and SHRpro:GUS , respectively . The constructs were verified by sequencing and were introduced into wild-type plants by Agrobacterium-mediated transformation . Primers used in cloning are shown in Table S3 . For GUS staining , plant tissues were incubated in the GUS assay solution ( 50 mM sodium phosphate buffer pH 7 , 5 mM Na2EDTA , 2 mM K3Fe ( CN ) 6 , 2 mM K4Fe ( CN ) 6 , 0 . 1% Triton X-100 and 0 . 04% X-Gluc ) at 37°C for 90 min for the DR5pro:GUS or for 3 h for the CYCB1;1:GUS , WOX5pro:GUS , and SHRpro:GUS leaf explants , and the stained tissues were then incubated in 70% alcohol at 37°C for 12 h . GUS staining was observed by using the SMZ1500 and Eclipse 80i microscopes ( Nikon , Japan ) . SEM analysis was performed as previously described [62] . For the Chromatin immunoprecipitation ( ChIP ) assay , leaf explants cultured at different time points on CIM or the 20 DAC calli were vacuum-infiltrated with formaldehyde crosslinking solution . ChIP was performed as previously described [29] , by using the antibody against H3 trimethyl-Lys 27 ( Upstate , USA , Cat . 07-449 ) . Results from real-time PCR represented the relative methylation levels , which were normalized against those produced by the primers for AG ( Figure S2 ) , whose values were arbitrarily fixed at 1 . 0 . For quantitative reverse transcription-polymerase chain reaction ( qRT-PCR ) , RNA extraction and reverse transcription were performed as described previously [62] , [63] . Three biological replicates were analyzed and each was tested by three technical replicates . Real-time PCR was performed using gene specific primers ( Table S3 ) and the results represented the relative expression levels , which were normalized against those produced by the primers for ACTIN , whose values were arbitrarily fixed at 1 . 0 . For microarray analysis , leaves from 20-day-old seedlings of wild type and 20 DAC calli were used for RNA preparation . Microarray was performed using the Affymetrix GeneChip system ( Affymetrix , USA , and Gene Tech Biotechnology Company Limited , Shanghai , China ) . Three independent experiments were performed , and expression of genes was considered significantly perturbed when the change was more than 2 . 0-fold with a P value<0 . 05 . The microarray data were deposited in the Gene Expression Omnibus ( GEO , http://www . ncbi . nlm . nih . gov/geo/ ) under the accession numbers GSE36479 and the analyzed data were shown in Table S2 . For ChIP-chip , about 100 ng DNA from leaves or calli was enriched , respectively , in the ChIP assay , and the ChIP-chip experiment was carried out by using Affymetrix high density tiling arrays according to the manufacturer's protocol ( Cat . 900594 , Affymetrix , USA , and Gene Tech Biotechnology Company Limited , Shanghai , China ) . ChIP-chip data were analyzed using the CisGenome software with standard procedures [64] , [65] . Genes with the significantly decreased H3K27me3 levels were defined as the peak cutoff MA ( Leaf-Callus ) ≥3 . 5 and MA ( leaf ) ≥2 . 5 , and those with the significant increased H3K27me3 levels were defined as MA ( Callus-Leaf ) ≥3 . 5 and MA ( Callus ) ≥2 . 5 . FDR ( <5% ) was settled as “Left ta” , which was recommended by the CisGenome software . The peak-gene association was settled from 3 , 000 bp upstream to the transcription start site ( TSS ) to the transcription end site ( TES ) . The database TAIR8 was used for annotation of the identified genes . For genes cluster based on the tissue specific expression , an online program Genevestigator ( www . genevestigator . com ) [30] was used . The ChIP-chip data were deposited in the GEO ( http://www . ncbi . nlm . nih . gov/geo/ ) under the accession numbers GSE34596 and the analyzed data are shown in Table S1 . | Callus formation is a necessary step in regenerating a new plant from detached plant tissues , and the nature of the callus is similar to that of the root meristem . In this study , we intended to address the molecular basis that directs different plant tissues to form the root-meristem-like callus . We found that leaves and cotyledons , but not roots , of PRC2 mutants curly leaf-50 swinger-1 and embryonic flower2 lost the ability to form a callus . Using ChIP–chip analysis , we identified genes that were changed markedly in the histone H3 lysine 27 trimethylation ( H3K27me3 ) levels during callus formation from leaf blades . Among these genes , a number of leaf-regulatory genes were repressed through PRC2-mediated H3K27me3 . Conversely , certain auxin pathway genes and many root-regulatory genes were derepressed through H3K27 demethylation . Our data indicate that genome-wide H3K27me3 reprogramming , through the PRC2-mediated H3K27me3 and the H3K27 demethylation pathways , is critical in directing cell fate transition . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"plant",
"science",
"plant",
"growth",
"and",
"development",
"plant",
"biology",
"genetics",
"epigenetics",
"biology",
"genetics",
"and",
"genomics",
"cell",
"fate",
"determination",
"histone",
"modification"
] | 2012 | Reprogramming of H3K27me3 Is Critical for Acquisition of Pluripotency from Cultured Arabidopsis Tissues |
Telomeres protect the chromosome ends from degradation and play crucial roles in cellular aging and disease . Recent studies have additionally found a correlation between psychological stress , telomere length , and health outcome in humans . However , studies have not yet explored the causal relationship between stress and telomere length , or the molecular mechanisms underlying that relationship . Using yeast as a model organism , we show that stresses may have very different outcomes: alcohol and acetic acid elongate telomeres , whereas caffeine and high temperatures shorten telomeres . Additional treatments , such as oxidative stress , show no effect . By combining genome-wide expression measurements with a systematic genetic screen , we identify the Rap1/Rif1 pathway as the central mediator of the telomeric response to environmental signals . These results demonstrate that telomere length can be manipulated , and that a carefully regulated homeostasis may become markedly deregulated in opposing directions in response to different environmental cues .
Telomeres are nucleoprotein structures located at the ends of chromosomes . Telomeres are essential for chromosome replication and stability [1] , and protect chromosome ends from degradation and deleterious chromosomal rearrangements [1] , [2] . In human embryonic cells , telomeres are elongated by the enzyme telomerase [3] . In somatic cells , however , telomerase expression is low , and telomeres shorten with each cell division due to the incomplete replication of the linear chromosome ends by conventional DNA polymerases . This progressive telomere shortening constitutes a “molecular clock” that underlies cellular aging [4] . Accordingly , telomere length is associated with cell senescence and longevity [5] , as well as with age-related disorders and cancer [6] . While short telomeres have been reported to predict early mortality [7] , recent work has shown that telomerase reactivation may reverse tissue degeneration in aged telomerase-deficient mice [8] . Three systematic genome-wide surveys in the yeast Saccharomyces cerevisiae [9]–[11] have revealed that mutations in at least 6% of the genes lead to alterations of telomere length . These TLM ( Telomere Length Maintenance ) genes span a broad range of functional categories and different cellular compartments . Integration of data from these large-scale mutant screens with information about protein–protein interactions has further permitted charting of the cellular sub-network underlying telomere length regulation in yeast [12] , [13] , revealing a complex set of interactions responsible for a very tight length homeostasis . Environmental stresses affect the regulation and the activity of many genes and accordingly may perturb telomere length homeostasis by altering the expression or activity of genes in the TLM network described above . Previous studies have suggested that emotional stress in humans is associated with telomere shortening , presumably through its effect on oxidative stress [14] , [15] . These studies , however , establish a correlation , but not causality . Here , we use controlled experimental approaches to explore a possible effect of the environment on yeast telomere length , and to identify the molecular mechanisms by which external signals exert their effect .
We exposed yeast cells ( S . cerevisiae ) to thirteen different environmental signals for 100–400 generations ( Figure 1 and Table S1 ) . Our results show that some stresses , such as high temperature , the addition of caffeine , and low levels of hydroxyurea resulted in telomere shortening , while others , such as added acetic acid and alcohols including ethanol , methanol , and isopropanol , caused a significant increase in telomere length ( Figure 1 ) . Strikingly , under alcohol stress telomeres were not only longer , but also exhibited length heterogeneity , indicating that the mechanism responsible for telomere length homeostasis , which preferentially elongates short , but not long telomeres [16] , was disrupted ( Figures 1 , 2 ) . The effect of alcohols on telomere length was independent of the ability of these cells to metabolize the alcohol: Upon ethanol treatment , isogenic petite yeast strains ( lacking mitochondrial function , and thus unable to utilize ethanol ) exhibited elongated telomeres ( Figure S1 ) . Importantly , however , many other environmental stresses , including oxidative stress , did not significantly alter telomere length ( Figure 1 and Table S1 ) , indicating that telomere length homeostasis is robust under many other environmental conditions . The effect of each stress on telomere length was concentration-dependent . In all cases , removal of the stressing agent resulted in a gradual restoration of wild type telomere length ( Figure 2A–C ) , demonstrating that the changes in telomere length were physiological rather than genetic , and thus may have been mediated by altered gene expression and protein activity . Under unperturbed conditions , telomere length can be modified either by disrupting the regulation of telomerase/telomere-associated nucleases or by recombination . To distinguish between these two mechanisms , we analyzed the response to stresses of cells unable to carry out homologous recombination due to a deletion of the RAD52 gene . rad52 cells responded to the stresses much as would a wild type strain , indicating that telomere length alteration in response to these stresses is not recombination-dependent ( Figure S2 ) and that the external signals affect telomerase or telomere-associated nucleases . To understand how external signals affect telomere length and to identify the mechanism behind this telomeric response to stress , we measured genome-wide transcript levels in yeast cells grown for 20 generations in the presence of stresses that showed an effect on telomere length ( ethanol , caffeine or high temperature ) , as well as in the presence of H2O2 , a stress that does not alter telomere length . The results were compared to genome-wide transcript levels of the same strain grown under standard conditions ( YEPD medium , 30°C ) . Using Significance Analysis of Microarrays ( SAM ) [17] with a false discovery rate ( FDR ) below 0 . 01 , we obtained a set of 1 , 744 , 1 , 404 , 1 , 670 and 1 , 019 differentially expressed genes for caffeine , 37°C , ethanol and H2O2 , respectively . General environmental stress responding ( ESR ) genes were not induced under these conditions , as expression level was measured after a long-term exposure to the stresses while ESR genes are induced for a short time period [18] . To identify the mechanisms responsible for telomere elongation and shortening , we sought genes that were differentially expressed only under shortening or only under elongating conditions ( Figure S3 ) . We integrated transcript abundance data with the known TLM network [13] that uses protein-protein interactions data , connecting TLM genes to the telomere maintenance machinery . The ( unweighted ) pairwise distances between stress-specific differentially expressed TLM genes were compared with pairwise distances of other TLM genes . This revealed that stress-specific , differentially expressed TLM genes lie significantly closer to each other for ethanol , caffeine and 37°C ( p<2E-33 , p<3E-27 and p<3E-50 , respectively ) , but not for hydrogen peroxide stress , which does not affect telomere length ( Materials and Methods ) . This phenomenon was unique to TLM genes under stresses that affect telomere length , suggesting that the differentially expressed TLM genes may be involved in transducing the external signals and disrupting telomere length homeostasis . Based on the analysis above , we generated a list of candidate genes for further analysis . Using strains from the yeast deletion library [19] and the DAmP library of hypomorphic mutants [20] we screened mutants in this list to identify genes important for telomere length maintenance under stress conditions . Strikingly , we found a strong correlation between the rate of change in telomere length and the initial length of the mutant: in ethanol , long tlm mutants elongate more rapidly than the wild type , while short tlm mutants elongate more slowly ( Pearson correlation , r = 0 . 61 , p<E-12 , Figure 3A ) . Similarly , in caffeine and at 37°C long tlm mutants shorten more rapidly , while short tlm mutants shorten more slowly than does the wild type ( Pearson correlation , r = −0 . 78 , p<2E-22 and r = −0 . 96 , p<9E-34 , respectively; Figure 3B–C ) . This correlation between abnormal telomere length and response magnitude to the stresses suggests that telomere elongation/shortening in the presence of external cues is carried out by the same basic mechanisms that maintain telomere length under unperturbed conditions . To identify the genes that mediate the telomeric response to stress and to understand how external signals are transduced to altering telomere length , we focused on mutants that disrupt this transduction and , therefore , show an atypical response to each stress ( Figure 3 ) . A remarkable such tlm mutant is rif1Δ , which exhibited a reduced response to ethanol and caffeine but normal response to 37°C ( Figures 3A and 4 ) , indicating that elongation by ethanol and shortening by caffeine are Rif1-dependent , while telomere shortening by high temperature relies on a different mechanism . The Rif1 and Rif2 proteins are negative regulators of telomerase that interact with the C-terminus of Rap1 , an essential protein that binds to the telomeric repeats [21] . Under normal growth conditions , short telomeres are preferentially elongated by a mechanism that depends on Rap1 . Mutations in the carboxy-terminus of RAP1 or down-regulation of the RAP1 gene lead to extreme telomere elongation and to an increase in telomere length variability , similar to what we observed in the presence of ethanol ( [22] , [23]; Figure 2 ) . Our transcript measurements detected a reduction in the level of Rap1 expression in cells grown in the presence of ethanol [ ( Table S2 ) ; and [24]] . These results suggest a model in which telomere elongation under ethanol stress is primarily due to reduced levels of Rap1 , which reduce Rif1 recruitment to telomeres . To test this hypothesis , we used a strain in which RAP1 was expressed from a Tetracycline-inducible promoter [25] . In this strain the level of Rap1 remained unchanged in the presence of ethanol ( Figure 4A ) and only a slight telomere elongation was observed ( Figure 4B ) . Also consistent with the model , a rap1-17 strain ( deleted for the C terminus of Rap1 ) , a rif1Δ single mutant and a rif1Δ rif2Δ double mutant exhibited attenuated responses to ethanol ( Figure 4B ) . Thus , the telomere elongation response to ethanol was abolished when a steady level of Rap1 protein was maintained or when Rif1 activity was eliminated , indicating that the Rap1- Rif1 pathway is central to telomere elongation in response to ethanol . Consistent with this hypothesis , chromatin immunoprecipitation ( ChIP ) experiments showed that upon exposure to ethanol there is a two-fold reduction in the level of Rif1 at telomeres , as well as a slighter reduction in the level of Rif2 ( Figure 4C ) . Since it is necessary for both elongation and shortening responses , Rif1 may play a general sensing/structural/regulatory role , rather than a catalytic one , in the telomeric response to environmental signals . This is consistent with recent studies that found a role for Rif1 in the regulation of chromatin structure and of DNA replication origin firing [26] , [27] . Remarkably , rif2Δ cells exhibited a strong response to ethanol ( Figure 3A ) , underscoring the different roles of Rif1 and Rif2 in telomere length maintenance [28]–[32] . We suggest that exposure to ethanol reduces the recruitment of the Rif proteins at the telomere ends , resulting in conditions permissive for indiscriminate telomerase recruitment , elongating both short and long telomeres , and yielding a broad distribution of telomere lengths ( Figure 2A ) . The insensitivity of rif1Δ mutants to ethanol could be due to the importance of Rif1p for the telomere elongation response , and/or the increased binding of Rif2 to telomeres in the absence of Rif1 . In agreement with this model , deletion of RIF2 caused over-extension of telomeres in ethanol ( Figure 3A ) ; a reduction of Rif1 telomere recruitment by ethanol in the strain deleted for RIF2 mimics a rif1Δ rif2Δ double mutant , which exhibits increased levels of telomere elongation . In contrast to these results , the RIF2 deletion had no effect on the reduction in telomere length upon exposure to caffeine or 37°C ( Figure 3B , C ) . Mutations in the TEL1 gene , which encodes the yeast ortholog of the mammalian ATM protein kinase , result in very short telomeres . Tel1 regulates the preferential elongation of short telomeres [33] by a pathway that also includes the MRX complex ( Mre11 , Rad50 , Xrs2; [34] ) . A separate regulatory branch includes the yeast Ku proteins [35] . Figure 3D shows that the tlm mutants with very short telomeres could be clearly separated into two groups: telomeres of mutants of the Tel1 pathway ( tel1Δ , mre11Δ , rad50Δ , xrs2Δ ) were hyper-responsive , while mutants of the NMD ( nonsense mediated decay , nmd2Δ , nam7Δ and upf3Δ ) and Ku pathways had only a mild response to ethanol . The fact that telomeres can be elongated by ethanol in the absence of Tel1 or of components of the MRX complex is surprising; notably , the wide size distribution observed upon exposure to ethanol ( Figures 1 , 2 ) , is consistent with a mechanism independent of the one that preferentially elongates the shortest telomeres , which depends on the Tel1 pathway [16] . The NMD pathway degrades mRNAs carrying nonsense mutations . In addition , it affects the steady state level of hundreds of mRNAs , including those known to act at telomeres ( e . g . , Est1 , Est2 , and two components of the CST telomeric capping complex , Stn1 and Ten1 [36] ) . Mutations in the NMD machinery lead to higher mRNA levels of these proteins and to short telomeres [37] . The NMD pathway has been recently shown to affect the fitness of cdc13-1 and yku70 mutants by controlling the expression of Stn1 , an essential telomere capping protein , which interacts with Cdc13 and participates in the recruitment of telomerase [38] . In nmd mutants , the response of telomeres to ethanol stress is reduced relative to wild-type strains , indicating that the NMD pathway is involved in telomere elongation during ethanol stress . We asked if upregulation of Ten1 and Stn1 is involved in this effect by overexpressing these genes in naïve cells and measuring the effect of ethanol on telomere length in these cells ( Figure S4 ) . Overexpression of Stn1 reduced the ethanol response and overexpression of both Stn1 and Ten1 completely abolished the telomere length response to ethanol . These results suggest that the level of CST activity , controlled by the NMD pathway , plays an important role in the telomere elongation response to ethanol . This is consistent with the proposed role of the CST complex in telomerase activation . Interestingly , mutations in the CST proteins are lethal when combined with a deletion of RIF1 [28]–[32] , indicating the existence of an essential overlapping function between the two telomere regulatory components . The roles of the CST and Rif1 in transducing the ethanol signal to the telomeres will be the subject of future research . Among the additional mutants with a reduced response to ethanol were doa4Δ , snf7Δ and did4Δ ( Figure 3A ) . DOA4 encodes an enzyme that removes ubiquitin from membrane proteins destined for vacuolar degradation . The Doa4 protein resides in the late endosome , where it interacts with the ESCRT-III machinery , which includes Did4 and Snf7 [39] . A role was previously observed for vacuolar traffic proteins in telomere length maintenance [40]; however , the precise mechanism remains enigmatic . Another mutant that shows apathy towards ethanol is hpr1Δ , defective for a component of the THO complex . Consistent with these results , mutations in HPR1 were recently shown to affect the expression levels of RIF1 [41] . In contrast to these genes , a deletion of HSP104 was hyper-responsive to ethanol . Hsp104 is a stress chaperone that plays an important role in maintaining prion particles in the cell [42] . It is unclear whether its role in telomere length regulation is related to its role in prion maintenance . Deletion of Rif1 and mutations in Rap1 also significantly decrease the telomeric response to caffeine , indicating that Rif1-Rap1 is not only involved in telomere elongation under ethanol stress , but also in telomere shortening under caffeine . Caffeine is a known inhibitor of phosphatydyl inositol-3 kinase related kinases ( PI3K-like kinases ) such as human ATR and ATM [43] and their yeast counterparts , Tel1 and Mec1 [44] . Therefore , we tested whether mutations in these target genes would abolish the telomere shortening caused by caffeine . Indeed , deletion of either TEL1 or MEC1 individually does not prevent the response to caffeine , but a double mutant tel1Δ mec1Δ is completely insensitive to the telomeric effect of caffeine ( Figure S5 ) , consistent with the known redundant function that these two kinases play in telomere biology [45] . Thus , caffeine causes telomere shortening by inhibiting the ATM/ATR-like regulatory kinases . Mutations in Rap1 and the deletion of Rif1 affect only the shortening rate in the presence of caffeine but do not affect the response to high temperature . High temperature has a broad , pleiotropic effect , and may alter telomere length via several mechanisms . Several TLM genes that , when mutated , result in short telomeres , are down regulated by high temperature ( Table S3 ) . However , no single deletion mutant failed to respond to high temperature by shortening its telomere length , suggesting that there are redundant functions among these responding genes . This result is consistent with a recent study [46] proposing that one or more telomerase components are intrinsically thermolabile . Accurate telomere length homeostasis is dependent on a large genetic network that includes ∼400 ( largely evolutionarily conserved ) genes [9]–[11] . Our results show that this network can be disrupted by several environmental signals , and by different regulation mechanisms that lead to altered telomere length . These responses are distinct from the stereotypic responses to stress [18] , and seem to be specific only to particular conditions . Telomere length and telomerase activity are important factors in the pathobiology of human disease . Age-related diseases and premature aging syndromes , for example , are characterized by the shortening of telomeres [47] . Tumor cells , on the other hand , prevent telomere shortening and telomere loss by up-regulating telomerase , thereby perpetuating cells with short telomeres and high chromosomal instability [48] . Thus , although the mechanisms at work differ , changes in telomere length fuel disease pathology in cancer and other premature aging syndromes . While previous studies have identified correlations between telomere length and environmental conditions such as mental stress [49] , socioeconomic status [50] , and health-related behavior in adults [51] , we extend those findings here by demonstrating direct causality between environmental cues and changes in telomere length . This identification of mechanisms by which external signals modify telomere length significantly advances our understanding of the complex interplay of genes and environment . More critically , however , these findings also point a future path to strategic manipulations of telomere length that may well have important therapeutic implications in the treatment of human disease .
All the yeast strains used in this study are derivatives of BY4741 ( MATa ura3Δ met15Δ leu2Δ his3Δ ) , unless otherwise specified . Mutants were obtained from the yeast deletion library [19] or from the DAmP library of hypomorphic alleles [20] . Strains carrying genes with tetracycline-inducible promoters were taken from the library described in [25] . Petite BY4741 derivatives were obtained by plating cells on YEPD plates containing ethidium bromide . Strains deleted for MEC1 , TEL1 and SML1 were in the MS71 background [52] ( kindly provided by T . Petes ) . Telomeric Southern blots were carried out as in [53] . PCR fragments containing telomeric sequences and a genomic region that hybridizes to two size marker bands ( 2044 and 779 bp ) were used as probes . The telomere length was measured with an in-house software ( TelQuant ) using the size marker bands as reference . Telomere length was ∼1250 bp in wt cells [composed of the sub-telomeric region ( ∼900 bp ) and the telomere repeats ( ∼350 bp ) ] . Stress levels were calibrated to reduce growth by 40%–60% . Cells were subjected to the various stresses by serial transfer growth: a single colony of BY4741 was grown in rich medium ( YEPD ) , and 5 µl were used to inoculate 5 ml cultures under the various stress conditions ( in triplicates ) . The cultures were grown ∼10 generations before being diluted ( 1∶1000 ) into fresh medium . We analyzed stress-induced RNA response for caffeine , temperature of 37°C , ethanol and hydrogen peroxide ( H2O2 ) , using Affymetrix GeneChip Yeast Genome 2 . 0 arrays . Transcript levels were measured for three independent cultures grown in the presence of the stress agent , and were compared to a control set comprised of four wild-type measurements . To obtain differentially expressed genes between the stress-induced response and the control measurements , we ( i ) employed the Robust Multi-array Average ( RMA ) method for normalization and summarization of the Affymetrix arrays [54]; ( ii ) filtered probes which had more than half of their detection calls marked as absent; and ( iii ) employed the Significance Analysis of Microarrays ( SAM ) [17] with false discovery rate ( FDR ) below 0 . 01 . Following these procedures , we obtained a set of 1 , 744 , 1 , 404 , 1 , 670 and 1 , 019 differentially expressed genes for caffeine , 37°C , ethanol and H2O2 , respectively . We used the un-weighted TLM-based network described in [13] , representing the most likely network connecting TLM genes to the telomere maintenance machinery . We next compared the pairwise shortest ( unweighted ) distances in the network between stress-specific differentially expressed TLM genes and other TLM genes , revealing that stress-specific differentially expressed TLMs for ethanol , caffeine and 37°C lie significantly closer to each other than other TLM genes ( Wilcoxon ranked sum test , p<2e−33 , p<3e−27 and p<3e−50 for ethanol , caffeine and 37°C stresses , respectively ) . Reassuringly , the hydrogen peroxide stress showed no significant difference between the two types of TLM genes . Last , using an assembled yeast protein-protein interaction network [13] , we verified that stress-specific differentially expressed TLM genes are significantly closer in this network than other stress-specific differentially expressed genes ( p<6e−9 for all stresses ) , verifying that closeness on the network is not a general property of differentially expressed genes . In an attempt to identify stress-response related genes , we examined the elongation or shortening of the telomere for each knockout gene in the absence or presence of the stress . The elongation/shortening of the telomere in the presence of the stress displayed a linear relation with the initial length of the telomere ( Pearson correlation coefficient between the two variables is ρ = −0 . 77 ( p<9e−25 ) , −0 . 95 ( p<2e−38 ) and 0 . 36 ( p<7e−6 ) for caffeine , 37°C and ethanol , respectively ) . In order to detect outliers , we performed a robust linear regression analysis . Following [55] , we assumed that the residuals follow a normal distribution and identified the outlier genes as the most extreme 5% ( 2 . 5% from each side ) . The computations were performed using Matlab . Chromatin immuno-precipitation ( ChIP ) was carried out by standard methods [56] . The association of Rif1-HA , and Rif2-HA with Y′-element telomeres was detected using Santa Cruz Mouse anti HA monoclonal IgG antibodies ( SC-7392 ) . Real-time PCR ( RT-PCR ) reactions were carried out using the following primers: Y′-element : 5′-GGCTTGATTTGGCAAACGTT-3′ , and 5′-GTGAACCGCTACCATCAGCAT-3′ . ARO1: 5′-GTCGTTACAAGGTGATGCC-3′ , and 5′- CGAAATAGCGGCAACAAC-3′ . The relative fold enrichment\depletion of the telomere-associated proteins Rif1 and Rif2 was calculated as follows: [telIP/ARO1IP]/[tel input/ARO1input] [57] . | Over 70 years ago , Barbara McClintock described telomeres and hypothesized about their role in protecting the integrity of chromosomes . Since then , scientists have shown that telomere length is highly regulated and associated with cell senescence and longevity , as well as with age-related disorders and cancer . Here , we show that despite their importance , the tight , highly complex regulation of telomeres may be disrupted by environmental cues , leading to changes in telomere length . We have introduced yeast cells to 13 different environmental stresses to show that some stresses directly alter telomere length . Our results indicate that alcohol and acetic acid elongate telomeres , while caffeine and high temperatures shorten telomeres . Using expression data , bioinformatics tools , and a large genetic screen , we explored the mechanisms responsible for the alterations of telomere length under several stress conditions . We identify Rap1 and Rif1 , central players in telomere length maintenance , as the central proteins directly affected by external cues that respond by altering telomere length . Because many human diseases are related to alterations in telomere length that fuel the disease's pathology , controlling telomere length by manipulating simple stressing agents may point the way to effective treatment , and will supply scientists with an additional tool to study the machinery responsible for telomere length homeostasis . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Environmental Stresses Disrupt Telomere Length Homeostasis |
We report efficacy and safety outcomes from a prospective case series of 31 late-stage T . b . gambiense sleeping sickness ( Human African Trypanosomiasis , HAT ) patients treated with a combination of nifurtimox and eflornithine ( N+E ) in Yumbe , northwest Uganda in 2002–2003 , following on a previously reported terminated trial in nearby Omugo , in which 17 patients received the combination under the same conditions . Eligible sequential late-stage patients received 400 mg/Kg/day eflornithine ( Ornidyl , Sanofi-Aventis ) for seven days plus 15 mg/Kg/day ( 20 mg for children <15 years old ) nifurtimox ( Lampit , Bayer AG ) for ten days . Efficacy ( primary outcome ) was monitored for 24 months post discharge . Clinical and laboratory adverse events ( secondary outcome ) were monitored during treatment . All 31 patients were discharged alive , but two died post-discharge of non-HAT and non-treatment causes , and one was lost to follow-up . Efficacy ranged from 90 . 3% to 100 . 0% according to analysis approach . Five patients experienced major adverse events during treatment , and neutropenia was common ( 9/31 patients ) . Combined with the previous group of 17 trial patients , this case series yields a group of 48 patients treated with N+E , among whom no deaths judged to be treatment- or HAT-related , no treatment terminations and no relapses have been noted , a very favourable outcome in the context of late-stage disease . N+E could be the most promising combination regimen available for sleeping sickness , and deserves further evaluation .
Treatment of late-stage ( meningo-encephalitic , stage 2 ) human African trypanosomiasis ( sleeping sickness , HAT ) due to Trypanosoma brucei gambiense currently relies on a meagre and very problematic drug armamentarium , consisting of; ( i ) melarsoprol , an old , extremely toxic[1] injectable , which in certain transmission foci is <70% effective[2] due to parasite resistance [3]; ( ii ) eflornithine , a safer[4] , [5] and efficacious[4] , [6] drug ( if given in a 14-day standard regimen[7] ) that must however be administered intravenously with 24 hour nursing care , placing an additional workload on the already fragile health systems in HAT-endemic areas; and ( iii ) nifurtimox , an oral drug originally intended for Chagas disease that is used on an off-label basis in HAT , but has shown disappointing cure rates as a monotherapy[8] . Since no new drugs are expected on the market for at least eight years[9] , combinations of these three drugs have long been considered the way forward to maximise cure rates , lengthen the drugs' lifespan by preventing further parasite resistance , and possibly improve safety and tolerability by reducing dosages of each partner drug , which would also result in easier administration . We previously reported[10] on a trial of the three possible combinations , melarsoprol-nifurtimox , melarsoprol-eflornithine , and nifurtimox-eflornithine ( N+E ) , initiated in 2001 in Omugo , Arua district , northwest Uganda , a historical HAT focus with high rates of melarsoprol failure[11] . The trial ( known as the Bi-Therapy Trial or BTT ) was interrupted on ethical grounds after 54 inclusions due to unacceptable death and severe adverse event rates in the melarsoprol-containing arms , and authorisation from the Uganda National Sleeping Sickness Control Programme to switch to eflornithine as first-line gambiense HAT therapy in Omugo , replacing melarsoprol . In this trial , the 17 patients randomised to N+E experienced significantly better outcomes: no fatalities or relapses , less frequent treatment interruptions , and fewer , milder adverse events[10] , clearly delineating a promising new therapeutic avenue . The trial was conducted within a long-established Médecins Sans Frontières ( MSF ) programme offering standardised clinical management , including a range of supportive therapies , the option of second-line treatments for cases of relapse , and high follow-up rates for discharged patients ( in HAT , efficacy control visits are conducted up to 24 months post treatment completion ) , thus reducing the risk to patients from unsafe or inefficacious experimental treatment . Furthermore , northwest Uganda was at the time politically stable , facilitating long-term research . The BTT provided evidence of the superiority of N+E in terms of safety . Although efficacy results were not yet available , we considered that further study of the N+E combination was warranted , and , given the favourable study conditions in northwest Uganda ( a rarity for HAT-endemic settings ) , decided in mid-2002 to implement a case series study of N+E with the objective of gathering more efficacy and safety data on this combination . We opted against a multi-arm trial due to the much-decreased HAT prevalence in the area and the imminent closure of MSF's programme , which would have jeopardised any attempt to reach sufficient power for statistical comparisons , especially since a non-inferiority design would have been needed to conclude N+E was at least as safe and efficacious as the current perceived best option , 14-day eflornithine monotherapy . In this paper , we report patient outcomes from this follow-up study ( conducted between September 2002 and March 2005 , and known as the NECS , or Nifurtimox-Eflornithine Case Series ) , and provide a joint analysis of NECS and BTT patients treated with N+E , to our knowledge the very first to receive this combination in a research setting .
In mid-2002 , the centre of HAT screening activities shifted from Omugo , Arua district , to Yumbe District hospital , about 40 Km northeast . Accordingly , the NECS study recruited patients presenting to Yumbe hospital , on a sequential basis . Yumbe district ( pop . 253 000 , 2002 census ) borders Sudan . Most of the district's villages ( rural communities scattered over thinly forested savannah ) reported HAT cases in the decade prior to the study ( MSF , unpublished observations ) . We initially set a sample size of 153 , enough to detect a cure rate of 90% with ±5% precision and 10% predicted incomplete follow-up . However , the very low rate of passive HAT case detection , and unexpectedly low HAT prevalence detected in active screening sessions around Yumbe in late 2002 , soon made this target unlikely , a frequent problem in HAT studies due to the rapid decline of transmission in sites where control activities are implemented . We therefore decided pragmatically to carry on recruitment until February 2003 , corresponding with MSF's departure . For consistency purposes , we replicated exactly the BTT trial methods , detailed by Priotto et al . [10] . Briefly , non-pregnant patients with bodyweight >10 Kg and late-stage T . b . gambiense HAT , defined as microscopic evidence of infection in the cerebrospinal fluid ( CSF ) or a CSF total leukocyte count of >5/µL with trypanosomes detected in blood or lymph node fluid , were invited to participate in the study if they had no history of HAT treatment in the prior 24 months and if their follow-up could be ensured . After systematic deworming and antimalarial treatment , patients received 400 mg/Kg/day eflornithine ( Ornidyl , Sanofi-Aventis ) for seven days in six-hourly slow infusions , plus 15 mg/Kg/day ( raised to 20 mg for children <15 years old ) nifurtimox ( Lampit , Bayer AG ) for ten days in three daily oral doses . Clinically apparent adverse events were monitored on a daily basis until discharge , and graded using Common Toxicity Criteria[12] . Further safety measurements at baseline ( day 0 ) and discharge ( day 11 ) included haemoglobin measurement , thrombocyte and leukocyte ( total and differential ) counts , as well as hepatic ( alanine transaminase [ALT] , bilirubin ) and renal ( creatinine ) function indicators ( not done in BTT trial ) measured from serum by standard spectrophotometry . Parasitology was repeated using standard techniques at discharge and 6 , 12 and 24 months thereafter , and defaulters were traced . Efficacy endpoints ( the primary outcome ) were failure in case of ( i ) death within 30 days post treatment initiation or later if judged compatible with HAT , or ( ii ) relapse within the 24 months of follow-up based on parasites' reappearance in any body fluid , or an increasing CSF leukocyte count[10]; and cure for all other patients followed up to and including the 24 month visit . Safety endpoints ( secondary outcome ) consisted of the occurrence of adverse events in temporal association with treatment , including abnormal laboratory values . Anaemia was defined as hemoglobin <13 g/dL ( males ) or <11 g/dL ( females ) having decreased by >20% from baseline; leukopenia as <4000 leukocytes/mL and decreased by >30% from baseline; neutropenia as <2000 neutrophils/mL and decreased by >30%; bilirubin abnormality as >17 µmol/L bilirubin and increased by >1 . 5 fold; ALT abnormality as >12 UI/L ALT and increased by >2 . 5 fold; and creatinine abnormality as >97 µmol/L ( males ) or >80 µmol/L ( females ) creatinine and increased by >1 . 5 fold . Analysis was done on Stata 9 . 0 software ( Stata Corporation , College Station , Texas ) . For comparability reasons , we adopted efficacy estimate approaches used in previous HAT trials[13] , [14]: ( i ) by intention-to treat ( ITT: relapses and all deaths irrespective of cause considered failures; lost to follow-up considered cured if they were seen at least once and had not relapsed at the last visit ) ; ( ii ) per-protocol ( only patients meeting all evaluability criteria retained for analysis; only relapses and HAT-related deaths considered failures ) ; and ( iii ) a worst-case intention-to-treat scenario in which all relapses , deaths and losses to follow-up are considered failures . Below we provide findings for the NECS , the previously published BTT N+E arm , and both groups combined . We abstained from significance testing due to the low numbers in each group . Both the BTT and the NECS studies received ethical clearance from the Uganda National Council for Science and Technology , and all participants ( or their legal guardians ) provided written informed consent . ClinicalTrials . gov registration numbers are NCT00330148 ( BTT ) and NCT00489658 ( NECS ) .
NECS recruitment lasted from October 2002 to February 2003 . Out of 56 stage 2 HAT cases presenting to the hospital , 31 were eligible and included . Baseline characteristics ( Table 1 ) were broadly consistent with other stage 2 patients seen in Yumbe during MSF's intervention ( data not shown ) . NECS patients ( n = 31 ) compared well with the BTT group ( n = 17 ) , but a greater proportion had a low body mass index , they had a higher geometric mean CSF leukocyte count ( 118/µL versus 46/µL ) , and typical sleeping sickness markers ( insomnia , somnolence and psychiatric signs ) appeared more frequent , suggesting that NECS patients were on average more clinically advanced . All 31 NECS patients were discharged alive . No deaths within 30 days of treatment start occurred , but two patients died before the 6 month follow-up visit . A five-year old patient moved to Sudan after the 6 month visit ( during which he was found healthy and parasite-free ) , and was subsequently lost to follow-up . The remaining 28 patients completed all follow-up visits by March 2005 ( Table 2 ) . No relapses were detected ( Table 2 ) . In the main ITT analysis , the two deaths during follow-up were considered failures , while all other patients were considered cured , giving an efficacy of 29/31 or 93 . 5% ( Table 2 ) . Because the two deaths were not HAT- or treatment-related ( see details below ) , per-protocol analysis considers all patients as cured ( efficacy 31/31 or 100% ) , while the ITT worst-case scenario includes the patient lost to follow-up among the failures , yielding an efficacy of 28/31 or 90 . 3% . Corresponding efficacy estimates for the entire BTT+NECS series ( n = 48 ) , ranging from 91 . 7% to 100 . 0% , are shown in Table 2 . The first dead patient , a 35 year old male , was attacked five months after discharge while on a business trip to Sudan , and died of his wounds at home shortly thereafter ( he was reportedly healthy before his trip ) . The second , a 37 year old female , had been treated successfully for stage 1 HAT in Omugo in March 2000 ( 28 months before study enrolment ) , but was lost to follow-up after one year . She experienced episodes of psychosis in April 2002 , and was enrolled in the study in November . During treatment she had generalized seizures and fever , hallucinations , and drowsiness . Tracing staff performing a post-discharge home visit found that she was being kept in chains by her family , allegedly due to her psychotic behaviour; they also noted a wound on her lower leg , reportedly caused by her imprisonment . After inpatient treatment the wound improved , but once back home she was again chained , the scar became infected and led to cellulitis; her family did not bring her back for treatment and left her to die of suspected septicaemia in February 2003 . On balance , this death is unlikely to be related to treatment ( the patient was perfused in the arms; deep tissue infections have been reported during eflornithine infusions[5] , but this cellulitis case was secondary to a deep wound , and due to her family's neglect ) . Her apparent psychosis may be a HAT sequela , but pre-dates her participation in the study . The ratio of adverse events per patient was higher in the NECS study than in the BTT: 4 . 0 ( 125/31 ) versus 2 . 1 ( 36/17 ) . Ascertainment of tremors , dizziness and vomiting/nausea events , none of which major , was strikingly higher in the NECS , and largely explained this difference ( Table 3 ) . On the other hand , the ratio of major ( Common Toxicity Criteria intensity 3 or 4 ) events per patient was similar ( 0 . 2 [7/31] versus 0 . 3 [5/17] ) , and only one seizure was noted during the NECS treatment period , compared to four in the BTT . All adverse events resolved without ascertainable sequelae , and did not lead to any treatment interruption in the NECS study , thus yielding only one temporary suspension in the entire NECS+BTT cohort ( Table 3 ) . No biochemical adverse events were noted apart from one mild case of raised creatinine at the end of the treatment course ( 1 . 6-fold increase from baseline level ) . As spectrophotometry procedures were only optimised mid-way through the study , reliable ALT and bilirubin results are available for only the last 13 NECS patients . Bilirubin levels appeared to decrease slightly ( median 1 . 0 mg/dL at baseline versus 0 . 8 mg/dL at discharge ) and ALT remained constant ( at 8 UI/L ) . As in the BTT , post-treatment neutropaenia was a common occurrence affecting 9/31 ( 29 . 0% ) of patients , of which one had a major episode ( count<1000/µL; Table 3 ) . These patients were monitored clinically via home visits , but not re-tested after discharge .
The BTT and NECS studies represent the first experience with a nifurtimox and eflornithine combination within a research context . Though small and inconclusive , we believe these studies represent a ‘proof of concept’ justifying further N+E experimentation . Altogether , our group of 48 patients shows very promising results: a favourable safety profile within the context of HAT , only one temporary regimen interruption ( in the BTT ) , no treatment- or HAT-associated deaths , and no relapses in a setting where melarsoprol failures exceed 30% . By comparison , case-fatality rates among non-relapsing patients treated with melarsoprol were 4% ( 66/1596 ) in Omugo during 1996–2002 , and 1% ( 1/93 ) in Yumbe during 2000–2002 ( MSF unpublished data ) . Among adverse events , the high frequency of neutropaenia , already highlighted during the BTT , was confirmed as a safety concern in the NECS , although only one of the nine episodes was considered major . In the two weeks following treatment , this could lead to opportunistic infections that might not receive appropriate treatment in remote HAT foci , especially among very advanced stage 2 patients , who may already be immuno-compromised , and among HIV-positives . Haematological and post-discharge monitoring of future N+E cohorts is thus warranted . Because the N+E regimen we applied reduces the eflornithine dose by half , it might nonetheless carry a lower risk of neutropaenia than eflornithine monotherapy , unless a drug interaction exists . Neutropenia as well as other bone marrow toxic effects affect 25–50% of patients receiving eflornithine monotherapy[6] , but the numbers followed to date are small , and the clinical significance of such adverse events has not been studied in the context of HAT . Apart from the BTT , only two other published studies of HAT combinations are available: a short-course regimen of eflornithine plus melarsoprol had an efficacy of 93% for relapsing cases , with two deaths among 40 patients[15] , while in a melarsoprol-sensitive focus of the Democratic Republic of Congo , among 69 patients receiving a combination of melarsoprol and nifurtimox as part of an equivalence trial , 14 treatment interruptions ( 12 suspensions and two terminations ) , three deaths during treatment , and no relapses were noted[8] . The latter trial also showed that even a decreased dose of melarsoprol entailed a serious risk of life-threatening encephalopathy , highlighting the urgent need to discontinue first-line use of this toxic drug . Limitations of the NECS study include the lack of a comparator group , and possible ascertainment bias as regards certain mild ( intensity 1 and 2 ) clinical adverse events , leading to an over-estimation of their frequency ( moreover , under-ascertainment in the BTT trial may also be possible ) . BTT patients were enrolled in Omugo , and NECS patients in Yumbe: while these sites are nearby ( about 40 Km ) and of similar environments , the two patient groups may be systematically different . Baseline characteristics of the two groups were broadly comparable , but NECS patients may have been slightly more advanced ( Table 1 ) . Because HAT is a very focalised disease , circulating strains in either site may have been more virulent or more drug-resistant . Therefore , interpretation of data for the entire BTT+NECS cohort merits caution . Clearly , the adoption of N+E as first-line treatment cannot be predicated on the basis of these findings alone , although it could be a viable alternative for relapsing cases , on a compassionate basis . It is a sad reality in HAT that the paucity of research funds and interest , compounded by arduous field research conditions , have thus far led to very unconventional drug development pathways , with little in the way of in vitro experimentation and dosage optimisation , and very few completed , sufficiently powered and comparative trials . The future looks somewhat brighter . In 2003 , a Good Clinical Practice equivalence trial of N+E versus standard eflornithine was begun in the Republic of Congo by Epicentre and MSF; the so-called NECT ( Nifurtimox-Eflornithine Combination Trial ) has since become a multi-partner effort involving five other enrolment sites , including two in Uganda . Most sites are currently in the patient follow-up phase , and final results are expected in 2008–2009 . Meanwhile , several interesting molecules are being considered for stage 2 drug development[16] . As the prospects of HAT elimination from Africa look unrealistic , such research is greatly needed , and must receive greater attention from researchers , funding agencies , and governments of both rich and HAT-endemic countries . | African sleeping sickness ( Human African Trypanosomiasis , or HAT ) , due to the parasite Trypanosoma brucei gambiense , threatens millions across remote and conflict-affected regions of sub-Saharan Africa , and causes about 15 000 reported cases every year . Untreated HAT progresses from stage 1 ( infection of the blood and lymph ) to stage 2 ( invasion of the central nervous system ) , and ultimately death . Drugs for stage 2 are few . The historical mainstay , melarsoprol , is highly toxic and inefficacious in some areas due to parasite resistance . Eflornithine is the only viable alternative , already established as safe and efficacious , but difficult to administer and at risk of resistance if used in monotherapy . This paper reports on a series of 48 Ugandan patients treated with a novel combination of nifurtimox ( a drug registered for Chagas disease ) and eflornithine , 17 as part of a terminated trial , and 31 in a subsequent case series study . Despite the low sample size , findings are promising: no cases of treatment failure , no treatment terminations , and no HAT- or treatment-related deaths . Nifurtimox plus eflornithine may be the best treatment hope for stage 2 HAT patients in the next decade , while new drugs are developed . A larger , multi-centric trial of the combination is ongoing . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"pharmacology/drug",
"resistance",
"pharmacology/drug",
"development",
"infectious",
"diseases/protozoal",
"infections",
"infectious",
... | 2007 | Nifurtimox plus Eflornithine for Late-Stage Sleeping Sickness in Uganda: A Case Series |
Cardiac aging is a complex process , which is influenced by both environmental and genetic factors . Deciphering the mechanisms involved in heart senescence therefore requires identifying the molecular pathways that are affected by age in controlled environmental and genetic conditions . We describe a functional genomic investigation of the genetic control of cardiac senescence in Drosophila . Molecular signatures of heart aging were identified by differential transcriptome analysis followed by a detailed bio-informatic analysis . This approach implicated the JNK/dJun pathway and the transcription factor Vri/dNFIL3 in the transcription regulatory network involved in cardiac senescence and suggested the possible involvement of oxidative stress ( OS ) in the aging process . To validate these predictions , we developed a new in vivo assay to analyze heart performance in various contexts of adult heart-specific gene overexpression and inactivation . We demonstrate that , as in mammals , OS plays a central role in cardiac senescence , and we show that pharmacological interventions impinging on OS slow heart senescence . These observations strengthen the idea that cardiac aging is controlled by evolutionarily conserved mechanisms , further validating Drosophila as a model to study cardiac senescence . In addition , we demonstrate that Vri , the ortholog of the vertebrate NFIL3/E4B4 transcription factor , is a major genetic regulator of cardiac aging . Vri overexpression leads to major heart dysfunctions , but its loss of function significantly reduces age-related cardiac dysfunctions . Furthermore , we unambiguously show that the JNK/AP1 pathway , the role of which in cardiac aging in mammals is controversial , is activated during cardiac aging and has a detrimental effect on cardiac senescence . This data-driven functional genomic analysis therefore led to the identification of key components of the Gene Regulatory Network of cardiac aging in Drosophila and may prompt to investigate the involvement of their counterparts in the cardiac aging process in mammals .
Age-associated changes in cardiac structure and function may contribute to the markedly increased risk for cardiovascular disease , what urges the need to understand heart aging at the molecular level . Cardiac senescence is a complex process , characterized by impaired cardio-acceleration during stress ( cardiac reserve ) and increased arrhythmia; it involves interactions between age , lifestyle , various diseases and genetic components [1] . Addressing the genetic basis of heart aging in mammalian models is challenging , due to their long life span , the complexity of the process ( many genes may be involved ) , the complexity of the models themselves ( genetic redundancy ) and the complex interactions between genetic traits and age/disease/lifestyle . There is therefore a need for simpler model organisms where genetic components of cardiac senescence can be more readily and rapidly assessed . Drosophila has recently emerged as an attractive model of cardiac aging [2] . Drosophila develops to adulthood quickly , has a relatively short life span ( 50 to 70 days at 25°C ) and is a highly tractable genetic system . The feasibility of analyzing large cohorts of individuals in tightly controlled environments makes it a powerful model for aging studies . Analogous to observations in elderly humans , the maximal heart rate is significantly lower in old than young Drosophila [3] , and an age-associated increase in rhythm disturbances has been observed . Wessells et al [4] demonstrated that modulating the insulin signaling pathway ( which has a conserved role in regulating life span [5] ) exclusively in the fly heart prevents the decline in cardiac performance with age . Several recent reports further support the use of Drosophila for analyzing cardiac responses to age-related stress ( reviewed in [2] ) . For example , the K+ATP channel encoded by the dSur gene appears to play a protective role against hypoxic stress , very much like its vertebrate counterpart , and dSur expression decreases with age , consistent with it being involved in the heart aging phenotype [6] . However , in spite of these pioneering studies , the molecular pathways involved in the progressive modifications of heart performance with age , and their genetic and environmental control , are still poorly defined . Here , we describe a functional genomic approach to investigate the genetic control of cardiac senescence in Drosophila . First , we identified molecular signatures of heart aging by differential transcriptome analysis of young ( 10 days ) and aging ( 40 days ) adult hearts . Data mining , which included prediction of transcription regulatory networks [7] , suggested that the JNK/AP1 pathway and the Vri/NFIL3 transcription factors are central to regulating cardiac senescence . This analysis also identified a potential role of oxidative stress ( OS ) . These predictions were tested by analyzing measures of heart senescence in vivo in flies following heart-specific genetic manipulations . In particular , the JNK/AP1 pathway was clearly found to be activated during cardiac aging and to have a detrimental effect on cardiac senescence . Furthermore , we demonstrate that the transcription factor Vri/dNFIL3 is a major genetic regulator of cardiac aging . Vri overexpression led to major heart dysfunction and its loss of function significantly reduced cardiac senescence . Thus , our study reveals two major genetic determinants of Drosophila cardiac aging , and paves the way for further studies in mammals .
To identify molecular signatures of cardiac aging , we performed a differential transcriptome analysis to compare young ( 10 days ) and aging ( 40 days ) hearts . At age 40 days , manifestations of cardiac senescence are clearly visible [3] , [4] , although flies still exhibit low mortality ( <10% ) in all genetic conditions tested . Flies were raised in tightly controlled conditions and RNA extracted from dissected hearts was used to probe microarrays ( see experimental procedures for details ) : 3097 probes representing 1107 unique Drosophila genes were found to be differentially expressed between the two time points , including 635 genes induced and 472 repressed at age 40 days ( Table 1 and Table S1 and S2 , full transcriptome data are accessible in the Gene Expression Omnibus database under the accession number GSE40253 ) . To validate the microarray expression data by quantitative real-time PCR , seven genes with different expression profiles ( 3 down regulated and 4 up regulated ) were selected . In all cases tested , the changes observed in the arrays were confirmed ( Figure S1 ) . Almost half of the genes identified as being differentially expressed ( 523 , 47% ) have previously been described as age responsive in previous studies focused on whole individual , or other body parts ( Table S3 ) ; these genes therefore contribute to molecular signatures common to aging . However , more than half of the genes that were differentially expressed with age in the heart had not previously been identified as being age related . This indicates that the aging of the heart , and probably other organs , may present particular traits and confirms the relevance of tissue-specific transcriptome analysis . Differentially expressed genes were analyzed for Gene Ontology ( GO ) term enrichment ( Table 1 and Table S4 ) . The set of under-expressed genes ( cluster 1 ) was enriched for genes involved in cellular respiration and mitochondrial bioenergetics ( Table S5 ) , including those encoding mitochondrial proteins involved in electron transport . This appears to be a general signature of aging tissues in fly , mouse and humans [8] . The set of genes induced during cardiac aging ( cluster 2 ) was enriched for inflammation and immune defense genes , another general trend of aging tissue in flies [9] , [10] and mice [11] . Genes involved in carbohydrate metabolism also show significant changes during aging , indicating a putative modification of energy metabolism . Increased oxidative stress ( OS ) has repeatedly been linked to the aging process ( see [12] for recent review ) and the down-regulation of mitochondria-related processes observed here ( Table S2 and S5 ) may signal impaired mitochondrial function in the aging heart . We therefore compared heart-deregulated genes with published transcriptome analyses of oxidative stress-responsive genes in the adult fly [13] [14] . One-hundred-and-forty up-regulated genes in the aged heart were found to be also up-regulated in mild MnSOD overexpression conditions ( [13] , enrichment , p<10−20 ) that induce a mild H2O2 -mediated OS ( Table 1 and Table S6 ) . In addition , 63 up-regulated genes were also found to be activated following oxidative stress induced by paraquat treatment ( [14]; enrichment p<5 10−5 , Table S7 ) . These observations suggest that oxidative stress plays an important role in Drosophila heart aging . This notion was further supported by cardiac-specific manipulation of Catalase expression , a reactive oxygen species-scavenging enzyme ( see below ) . Genes differentially expressed with age may be regulated by transcription factors ( TFs ) which would constitute good candidates for the control of heart senescence . To investigate the transcriptional regulation of age-regulated genes , we performed in silico predictions of TF-binding motifs potentially involved in the regulation of genes up- and down-regulated in the aging heart . We used the recently published cisTargetX method [7] , [15] . cisTargetX uses identification of clusters of TF-binding motifs across the entire genomes of 12 Drosophila species and ranking statistics , to determine significant associations between motifs and subsets of co-expressed genes . It uses a large sequence space ( 5 kb up to the 5′ transcription start , and the introns of all genes ) and a comprehensive library of TF-binding motifs to predict potential regulatory motifs . We reasoned that it might therefore allow regulatory motif predictions without prior knowledge of the TFs involved . Motifs known to bind AP1 –the Fos/Jun heterodimer— were substantially enriched in the subset of over-expressed genes ( z-score = 6 . 30 ) defining a set of 98 potential AP1 targets ( Table 1 and Table S8 ) . Interestingly , AP1 is an established effector of the Jun N-terminal Kinase ( JNK ) signaling pathway , which is one of the most versatile stress sensors in metazoans and has been linked to aging in fly [16] . The JNK pathway in Drosophila involves Basket ( Bsk , dJNK ) and an additional kinase , Hemipterous , which serves to phosphorylate Bsk . Consistent with a role for the JNK pathway in cardiac aging , 22% ( 29/133 , p<10−7 ) of JNK targets previously identified in Drosophila S2 cells following either bsk or Hep inactivation [17] , [18] are present among the genes up-regulated at 40 days ( Table 1 and Table S9 ) . Interestingly , Jra , the Jun Drosophila orthologue , is one of these genes , suggesting a positive feedback loop of AP1 activation in the aging heart . These various findings strongly suggest that the JNK-AP1 pathway is involved in Drosophila heart aging . cisTargetX also identified Evi-1 motifs as being potentially involved in the regulation of over-expressed genes during Drosophila heart aging . Evi-1 motifs –characterized as binding the vertebrate zinc finger TF Evi-1— were greatly enriched ( z = 6 . 1 ) in the conserved non-coding sequences of this gene set ( Table 1 and Table S8 ) . A list of 30 genes potentially regulated by these motifs during heart aging is given in Table S8 . However , the corresponding drosophila TF has not been unambiguously identified: the drosophila genome contains two genes encoding Evi-1 homologues –hamlet ( ham ) , which regulates cell identity in the peripheral nervous system [19] and CG10348 of unknown function- and their binding motifs have not been characterized . In addition , Evi-1 motifs are constituted of tandem repeats of GATA motifs , such that their significant association with up-regulated genes may alternatively indicate the involvement of TFs of the GATA family . In contrast to the set of age-induced genes , we did not identify enrichment in AP1 or Evi-1 motifs in the set of down-regulated genes . However we found a strong enrichment ( z = 6 . 5 ) for motifs associated with the bZip TF Vrille ( Vri ) and its mammalian homologue NFIL3 ( also called E4BP4 ) , which act mainly as a transcriptional repressors [20] ( Table 1 and Table S8 ) . Both Vri and NFIL3 have been implicated in many developmental processes but Vri/NFIL3 has never previously been reported to be associated with aging either in vertebrates or in Drosophila . These in silico predictions of TF-binding motifs thus suggest that Vri , dJun and an un-identified TF able to bind Evi-1 motifs participate in the transcriptional regulatory network involved in heart senescence . We developed a new in vivo assay to analyze heart performance in various contexts of adult heart-specific gene overexpression and inactivation . Our aim was to set up an assay suitable for large-scale analysis , allowing conditional and tissue-specific genetic manipulations and measurements of heart senescence in physiological conditions in intact ( not dissected ) individuals . We exploited the GeneSwitch system [21] which allows expression to be induced by RU486 feeding . A heart-specific Geneswitch driver , Hand-GS , was constructed and combined with a UAS-mitoGFP construct . Hand-GS>UAS-mitoGFP flies fed with RU486 exhibit a cardiac fluorescence strong enough for high-speed video recording through the cuticle of anaesthetized flies of various ages ( Figure 1A ) . Flies imaged under UV light exhibit a heart rate ( HR ) in the same range as those obtained in previous studies on intact flies [22] , [23] , [24] . HR under UV is slightly higher and more regular than the mean HR obtained under visible light on anesthetized flies ( Figure S2 ) . Image acquisition and analysis of M-Modes are described in details in Text S1 and Figure S2 . Relevant functional measures were extracted from M-Mode analysis and used to quantify heart performance , including the mean time between successive end-diastolic positions ( Heart Period , HP ) , an Arrhythmicity Index ( AI ) defined as the standard deviation of the HP normalized to the median HP as described in [25] and the End-Diastolic Diameter ( EDD ) . We checked that heart performance was not dependent on RU concentration at any age ( comparison of two RU486 treatments −20 µg/ml and 100 µg/ml- was performed ) , and was not affected by expression of luciferase or a control ds-RNA ( Figure S3 ) . We observed a progressive increase in HP with age , mainly between age 10 days and 45 days , and an increased AI mainly between ages 45 days and 60 days ( Figure 1B , 1D and 1E , Video S1 , S2 and S3 ) . Similar age-related decreases in heart performance have been described in flies using other methods of heart beat detection , including semi-intact preparations , in which the heart is surgically exposed [3] , [4] , [25] . In addition , we observed a progressive decrease in End-Systolic Diameter ( Data not shown ) and End-Diastolic Diameter ( EDD , Figure 1C ) . EDD decreased by 13 , 4% between ages 10 and 45 days , and by 26 , 9% between ages 45 and 60 days . Our assay appears to have various advantages over previously used methods for studying cardiac aging . First , heart activity is monitored in intact flies , in optimized physiological conditions . Second , it does not require highly specialized equipment and is suitable for large-scale analysis since movie acquisition is fast ( 30 flies can be recorded in one hour ) and each step of the analysis is automated . Finally , the heart-specific inducible driver provides the opportunity to study adult-specific gene overexpression or ds-RNA mediated inactivation , independently of other developmental effects . The transcriptome analyses suggested that the fly heart is subject to rising levels of oxidative stress with age ( see above ) . This was confirmed by manipulating Catalase activity , an antioxidant enzyme which detoxify H2O2 , specifically in the cardiac tube at adulthood , and by analyzing the induced transcriptome modifications at 40 days ( Table 2 and Table S12 ) . Genes whose expression is induced following increased OS –eg induced in RNAi mediated Cat knockdown compared to Cat overexpression- are highly enriched in genes that are overexpressed in cardiac tubes at 40 days compared to 10 days ( cluster 2; 59/635 p = 4 . 9 10−35 ) and is enriched in genes participating in defense response ( Table S13 ) . On the contrary , the set of genes that are up regulated following reduced OS -eg in Cat overexpressing hearts compared to Cat knockdown- is highly significantly enriched in cluster 1 genes ( gene set downregulated at 40 days compared to 10 days , 119/472 , p = 3 . 22 10−68 ) . This analysis thus confirms the central role played by OS in the age related cardiac transcriptome dynamics . We therefore directly tested the involvement of oxidative stress in cardiac aging by genetically modulating the expression of the antioxidant enzymes Superoxide dismutase ( SOD ) which detoxify O2•– and Catalase in the heart and analyzed heart parameters . Catalase inactivation ( Cat-IR condition ) led to a strong age-dependent deleterious phenotype . HP was 71% higher in 60-day-old flies than controls of the same age ( Figure 2A and 2C , Video S4 , Table S11 ) . Inversely , over-expression of Catalase ( Cat OE condition ) substantially improved cardiac performance in old flies ( Figure 2A and 2D , Video S5 ) , with lower HP and arrhythmia in both 45 and 60 days old flies compared to controls . Expression of another enzyme with catalase activity ( Catalase B , CG9314 ) , normally not expressed in cardiac tissues , also improved cardiac function ( Figure S4A ) , confirming the beneficial effect of catalase activity for preventing cardiac senescence . There were no significant differences in heart performance , at any individual time-points , between flies with enhanced or decreased expression of SOD1 ( CuZn or cytosolic SOD ) ( Figure S4B and S4C ) , although the rate of AI increase as function of age was slightly higher in SOD1 depleted flies ( Table S11 ) . Altogether , this suggests that H2O2-mediated oxidative stress is predominant in the cardiac aging process . Next , we treated wild type flies with EUK-8 ( + EUK8 condition ) , a synthetic superoxide dismutase and catalase mimetic , from the age of 30 days , and compared heart performance between treated and untreated 45 and 60 day-old flies . EUK-8 improved both HP and AI ( Figure 2A , 2E , Video S6 and Table S11 ) . In particular , it fully abolished the age-dependent development of cardiac arrhythmia: the AI was not different in 10-day-old and EUK-8-treated 60 day-old flies ( p = 0 . 3 ) , whereas AI increased by 74% in control flies between ages 10 and 60 days ( p<5 . 10−3 ) . EUK-8 treatment also prevented the age-related EDD decrease ( Figure 2B ) : the EDD was not different ( p = 0 , 26 ) between 30 days ( start of treatment ) and 60 days old treated flies , whereas EDD decreased by 29% ( p<5 . 10−3 ) between these two ages in untreated flies . Similarly , flies overexpressing Catalase presented a stable EDD between ages 10 days and 60 days ( p = 0 , 13 ) . Interestingly , in mammals , expression of mCAT , a catalase artificially targeted to mitochondria , protects mice from cardiac aging [26] . EUK-8 or other synthetic SOD/Cat mimetic also appear to improve heart function in pathological contexts [27] [28] [29] . Our findings and these various observations highlight the role of oxidative stress in both pathological and normal cardiac aging and indicate that mechanisms of cardiac aging are conserved between flies and mammals . In addition , we show here that pharmacological antioxidant strategies can improve heart function in old individuals . The heart-specific inducible driver can be used to study adult-specific gene over-expression or ds-RNA-mediated inactivation , independently of other developmental effects . This allowed us to test the in silico predictions of a transcriptional regulatory network potentially involving dJun and Vri as regulators of cardiac senescence . A large number of JNK-responsive and potential AP1 target genes , as well as dJun itself , were up-regulated in the aging fly heart , suggesting that the JNK/AP1 pathway is involved in their transcriptional up regulation ( cluster 2 , see above ) . To confirm the functional involvement of dJun in the transcriptional deregulation of these up-regulated genes , we performed a transcriptome analysis of 40 days old hearts following dJun knockdown by dsRNA and focused on genes whose expression is repressed in dJun knockdown compared to control cardiac tubes ( Table 3 and Table S14 ) . Not surprisingly , the set of genes whose expression level is reduced by at least 1 . 5 fold following dJun knockdown is significantly enriched in JNK target genes ( 29/135 genes , p = 1 . 1 10−5 ) . Importantly , this gene set was also highly significantly enriched in cluster 2 genes ( 138/635 genes , p = 10−19 ) , what support a function for dJun in the age-related transcriptome deregulation reported above . In addition , the predicted AP1 targets within cluster 2 ( Table S8 ) were also enriched in this gene set: almost 25% of predicted cistargetX AP1 targets are under-expressed after dJun knockdown ( 23/99 , p = 1 . 8 10 −5 ) . Altogether , these data validate the cisTargetX predictions made from cluster 2 and confirm the central role played by the JNK/AP1 pathway in the up regulation of a significant proportion of genes in aging hearts . We therefore used ds-RNA-mediated inactivation to abolish dJun activity in the heart ( dJun-IR condition ) and measured heart parameters . This significantly improved heart function and led to a higher heart rate and less arrhythmia ( Figure 3A and 3C , Video S7 ) , in both young ( 10-day-old ) and aged ( 45-day-old ) flies than in the corresponding controls . HP and AI were respectively 26 . 5% and 37 . 6% lower in 45-day-old flies than in controls of the same age . Noticeably , HP and AI were similar in 3 days old dJun-IR and control flies . In addition to the beneficial effects already observed in 10 days old flies , dJun partial inactivation significantly slowed down the rate of HP increase as a function of age ( Table S11 ) and prevented the age-related EDD decrease ( Figure 3B ) . We have tested an independent RNAi line targeting different sequences of dJun and observed similar beneficial effects on HP , AI and EDD in 45 days old flies ( Figure S5A ) . In addition , RNAi mediated inactivation of kayak ( the Drosophila Fos homologue ) also significantly improved AI and EDD in old flies ( Figure S5A ) , confirming the central role of the Fos/Jun heterodimer AP1 in cardiac aging . We also analyzed the effect of dJun overexpression on heart function ( dJun OE condition , Figure S5B ) . We observed an increased HP in 3 days old and 10 days old flies ( respectively 23% and 10% ) . In older flies , dJun overexpression did not affect HP , probably because the JNK pathway was already activated . JNK may be activated in the fly heart through OS signaling since ROS are potent activators of JNK by several mechanisms ( reviewed [30] ) . However , JNK pathway target genes were not enriched following cat misexpression ( 1/133 genes , p = 0 . 53 , see Table 2 ) . Additional experiments are therefore required to determine the causes for JNK activation in aging hearts . The set of potential AP1 transcriptional targets that are up-regulated during aging is strongly enriched in genes involved in cytoskeleton organization and , in particular , in regulation of actin polymerization ( Table S10 ) . In addition , the set of genes that are deregulated following cardiac specific dJun knockdown at adulthood is enriched in genes involved in vesicle mediated transport ( Table S15 ) . This suggests that dJun activation in old flies may affect heart function by remodeling the actin network and/or by affecting vesicle mediated transport in cardiomyocytes; this notion should be explored in further studies . In mammals , the role of the JNK pathway in heart function and diseases ( including cardiac hypertrophy , ischemia/reperfusion injury and pathological remodeling ) is unclear and both in vitro and in vivo studies have given contradictory results ( recently reviewed in [31] ) . Here , we show unambiguously that the age-dependent activation of Jra/dJUN in the fly heart is detrimental rather than protective . JNK has numerous evolutionarily conserved downstream targets besides the AP1 family members Jun and Fos , notably Forkhead Box O ( FoxO ) proteins [32] , [33] , [34] . Interestingly , overexpression of dFoxo prevents the decline in cardiac performance in aging flies [4] . Combined with our results , this suggests that , at least in Drosophila , JNK plays a complex and dual role in heart aging that can be decoupled: positive effects mediated by the insulin signaling pathway ( IIS ) , and negative effects mediated by AP1 . Motifs associated with Vri/dNFIL3 were over-represented in genes down regulated with age . We confirmed that vri is able to regulate putative target genes identified by cisTargetX , by overexpressing vri and analyzing the expression of 3 of its putative targets by Q-PCR in 10 days old adults . As shown in Figure S6C , all 3 tested genes were repressed following vri overexpression , indicating that vri is a bona fide repressor of their expression . We therefore investigated its involvement in cardiac aging . Overexpression of vri ( vrille OE condition ) led to a large decrease in heart rate even by age 10 days . This effect was dependent on the level of Vri overexpression: 10 days old flies under RU486 10 µg/ml treatment did not exhibit increased HP , whereas under RU486 100 µg/ml treatment ( the RU486 concentration currently used in this study ) , HP was increased by 38% compared to controls ( Figure S6B ) . In 45 day-old flies , HP was 142% higher than in controls of the same age ( Figure 3A and 3E , Video S8 ) , whereas AI was not significantly different . This strong increase in HP was associated with a moderate dilated heart phenotype: between ages 10 and 45 days , EDD increased by 19% in flies overexpressing vri , whereas in control flies EDD decreased by 13% . Inversely , vri inactivation strongly improved cardiac performance in old flies ( vrille-IR condition , Figure 3A , 3B , and 3D; Video S9 and Table S11 ) . In control hearts , HP increased by 55% between ages 10 and 45 days , whereas it only increased by 6% when vri was inactivated . Inactivation of vri prevented EDD decrease ( EDD remained stable between ages 10 and 45 ) and significantly reduced arrhythmia: AI was 23 . 6% and 36% lower in 10-day-old and 45-day-old flies than controls of the same ages . We have tested an independent RNAi line targeting different sequences of vrille ( vrille-IR ( II ) ) and also observed beneficial effects on AI , HP and EDD ( Figure S6A ) . These data clearly establish that Vri/dNFIL3 is a major regulator of heart senescence in flies . Furthermore , our molecular analyses suggest that it acts mainly by repressing downstream targets . The set of potential Vri/NFIL3 transcriptional targets that were down-regulated during the aging process was strongly enriched in genes encoding mitochondrial proteins ( Table S10 ) and in particular those of the mitochondrial respiratory chain complexes . This transcriptional control has been confirmed by Q-PCR for CG11015 , a component of the complex V of the electron transport chain ( Figure S6C ) . This suggests that Vri may promote mitochondrial dysfunction in the aging heart through repression of genes encoding mitochondrial proteins . The effect of Vri on EDD is complex , since Vri inactivation prevented age-related EDD decrease , and Vri overexpression induced a dilated phenotype ( increased EDD ) . Whether these two phenotypes involved the same mechanism or relied on independent Vri downstream targets remains to be clarified . Interestingly , in humans , mitochondrial respiratory chain disorders are frequently associated with hypertrophic or dilated cardiomyopathies ( reviewed in [35] ) , suggesting that the dilated heart phenotype induced by Vri overexpression in flies may be due to strong mitochondrial respiratory defects . We also observed a transcriptional repression of prx5 by Vrille ( Figure S6C ) . prx5 encodes a peroxiredoxin localized in several cellular compartments including mitochondria , that confers resistance against oxidative stress and promotes longevity in Drosophila [36] . This demonstrates that Prx5 is a downstream target of Vri that may couple the Vri pathway with ROS levels during the cardiac aging process . Recently , NFIL3 has been shown to be required for correct zebrafish heart development [37] . However , the role of NFIL3 in adult heart function and aging has not been described , and investigating the role of NFIL3 in cardiac aging in mammals may be fruitful . Genetic redundancy and long lifespan make it difficult to investigate the genetic control of cardiac senescence in mammals . Our functional genomic approach in Drosophila allowed these problems to be avoided and led to the identification of key components of the gene regulatory network of cardiac aging . First , we demonstrate that , as suggested in mammals , oxidative stress plays a central role in fly cardiac senescence . Our observations therefore support the idea that mechanisms of cardiac senescence are evolutionary conserved . Second , although the role of the JNK pathway in age-related processes in mammals is controversial , we show here that the dJun TF is activated during cardiac aging and has a detrimental effect on cardiac performance . Third , the transcription factor Vri/dNFIL3 is shown to play a central role in the cardiac aging process . Given that Drosophila heart shares lots of common points with that of vertebrates , our findings suggest that Vri and dJun orthologues are putative targets for counteracting cardiac aging in mammals . To conclude , our work illustrates the relevance of data driven tissue-specific functional genomic analysis of the aging process in a model organism which allows straightforward genetic manipulations . Numerous studies , including transcriptome analysis in rodent models indicate that the effects of aging are in large part tissue-specific . Accordingly , half of the genes identified in our transcriptome analysis appear to be uniquely deregulated in the drosophila heart , suggesting cardiac specific mechanisms of aging . This is further supported by our observation that activation of the JNK/AP1 pathway is linked to accelerated heart senescence , while its activation organism-wide lead to increased life-span [38] . This probably indicates that activation of the JNK pathway has different outcomes with respects to senescence in different tissues or organs . As a matter of facts , its activation in the nervous system increases longevity [16] , but reduces lifespan when achieved in intestinal stem cells [39] . Identification of the downstream effectors of this pathway in these different tissues and comparison with the cardiac specific targets identified in the present study may allow shedding light on its age related tissue specific effects .
Cardiac transcriptome dynamics between 10 and 40 days: Adult males ( w1118 , Canton S isogenic line ) were grown in batches of 30 flies at 25°C and 60% hygrometry . Total RNA was extracted with Trizol and mRNA amplified using messageAmp ( Ambion ) . Labeled aRNA was hybridized with Nimblegene arrays . After normalization of expression levels , differentially expressed genes were determined using the Limma software package . Gene Ontology enrichments were assessed using Flymine ( www . flymine . org ) and Transcription Regulatory Networks studied using cisTargetX ( http://med . kuleuven . be/cme-mg/lng/cisTargetX/ ) . Cardiac transcriptome modifications induced by Catalase A Gain and Loss of Function: Adult males ( w;+/+; Hand-GS/UAS-Catalase-IR ( Cat knockdown ) and w;+/+; Hand-GS/UAS-Catalase ( Cat overexpression ) ) were grown in batches of 30 flies at 25°C and 60% hygrometry on food supplemented with 100 µg/ml of RU . RNA extraction was performed as above . Labeled aRNA were hybridized with Agilent arrays and differentially expressed genes determined using the Rank Product software within the TMev software suite . Cardiac transcriptome modifications induced by dJun RNAi mediated knockdown: Adult males ( w;+/+; Hand-GS/UAS-dJun-IR ) were grown in batches of 30 flies at 25°C and 60% hygrometry on food supplemented or not with 100 µg/ml of RU . RNA extraction , micro array hybridization and data analysis were performed as described in the case of Catalase gain and loss of function . Flies expressing the GFP protein targeted to the mitochondria ( mitoGFP ) to label the cardiac tube were anaesthetized with Triethylamine and observed under a Zeiss SteREO Lumar . V12 Stereomicroscope , with a NeoLumar S 1 . 5× objective . Video movies were acquired with an AxioCamHR Camera . M-Modes were generated by horizontal alignment of rows extracted at the same position from each movie frame by using ImageJ , with automated positioning of the acquisition zone . Cardiograms ( defined by the distance between the maxima of GFP fluorescence on each side of the median position of the heart at each time point ) were then generated from M-Modes using an image processing algorithm developed with Matlab R2010b . The temporal positions for each end-systolic and end-diastolic position of the heart were extracted by finding all local maxima and minima on the cardiogram . The resulting file was incorporated into an Access DataBase to extract , for each cardiogram , the Heart Period ( HP ) and the Arrhythmicity Index ( AI ) . Statistical significance was assessed by non-parametric Wilcoxon analysis [40] . Details of the Materials and Methods can be found in Text S1 . | Age-associated changes in cardiac structure and function have been implicated in the markedly increased risk of cardiovascular disease , but the molecular basis of these processes is ill-defined . It is difficult to study the genetics of heart aging in mammalian models because of their long life spans and their complexity , involving notably genetic redundancy . Here , we address this issue through identification of molecular signatures of cardiac aging in Drosophila , a model organism in which heart senescence occurs within 2 months . Tissue-specific transcriptome comparison of young and aging fly hearts were performed followed by in silico predictions of the regulatory networks involved . This analysis implicated oxidative stress ( OS ) , the JNK/dJun pathway , and Vri/dNFIL3 in the gene regulatory network that drives cardiac senescence . Measuring heart variables in vivo following heart-specific genetic and pharmacological manipulations confirmed these predictions . We show that OS has a central role in the aging of the fly heart . Moreover , heart-specific partial knockdown of dJun and Vri prevented cardiac senescence , demonstrating that they are essential regulators of cardiac aging . Thus , our results uncover two major genetic determinants of Drosophila cardiac aging whose activities enhance heart senescence . It may therefore be valuable to investigate their involvement in the cardiac aging process in mammals . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"functional",
"genomics",
"animal",
"genetics",
"animal",
"models",
"genome",
"analysis",
"tools",
"drosophila",
"melanogaster",
"model",
"organisms",
"gene",
"expression",
"regulatory",
"networks",
"biology",
"transcriptomes",
"microarrays",
"genetics",
"genomics",
"gene... | 2012 | dJun and Vri/dNFIL3 Are Major Regulators of Cardiac Aging in Drosophila |
Biomedical research has been previously reported to primarily focus on a minority of all known genes . Here , we demonstrate that these differences in attention can be explained , to a large extent , exclusively from a small set of identifiable chemical , physical , and biological properties of genes . Together with knowledge about homologous genes from model organisms , these features allow us to accurately predict the number of publications on individual human genes , the year of their first report , the levels of funding awarded by the National Institutes of Health ( NIH ) , and the development of drugs against disease-associated genes . By explicitly identifying the reasons for gene-specific bias and performing a meta-analysis of existing computational and experimental knowledge bases , we describe gene-specific strategies for the identification of important but hitherto ignored genes that can open novel directions for future investigation .
Recent studies have demonstrated the highly imbalanced research effort directed towards individual human protein-coding genes [1–8] , which manifests itself in several ways , including the number of publications per gene , the number of human-curated and computationally predicted functional annotations , the number of gene names and gene symbols , and the number of patents containing their nucleotide sequences ( S1 Fig ) . Plausibly , this observed disparity could reflect a lack of importance of many genes , but more likely it could also reflect existing social structures of research [9 , 10] , scientific and economic reward systems [11 , 12] , medical and societal relevance [13–15] , preceding discoveries [2 , 16] , serendipity [17 , 18] , the availability of technologies [19 , 20] and reagents [6 , 21] , and other intrinsic characteristics of genes [22–24] . It remains unclear , however , if any of these factors can significantly explain the observed number of publications on individual human genes . Nor is it known whether descriptions about the formation of scientific knowledge translate into gene-specific insight , and whether these reasons for historically grown bias could already be mitigated by current experimental possibilities . In order to address these challenges , we created a database cross-referencing chemical , physical , biological , historical , bibliometric , financial , technological , and experimental data on all human protein-coding genes from 36 different sources ( see Materials and methods ) . Using this resource , we show how characteristics of genes relate to the macroscopic output of biomedical research in terms of the number of publications , perceived biological importance of genes , funding , and translational activities . We show different examples of how this resource can be used to define strategies for a more efficient exploration of the space of biological functions , and provide high-level gene-specific analyses in a series of supplementary tables .
To test if measurable intrinsic chemical , physical , and biological features of genes and gene products alone suffice to describe the number of publications per gene , we gathered 430 features per gene , which could either be computed from known sequences of these genes or obtained from previously published genome-scale experiments ( Fig 1A ) . Intriguingly , we observed that 33% of the protein-coding genes carrying an official gene name had an incomplete catalog of features . The dominant reasons for the absence of features were the absence of reported insertions within recent Clustered Regularly Interspaced Short Palindromic Repeats ( CRISPR ) loss-of-function screens ( about 13% of genes , depending on assay ) , the absence of detectable RNA across all tissues and cell lines surveyed by the human protein atlas ( 6% of genes ) , the absence of validated RNA molecules within the Genbank reference database of RNA molecules ( 5% of genes ) , and the absence of reported protein molecules within the UniProt reference database for protein molecules ( 3% of genes ) ( S2A Fig , S1 Table ) . Foreshadowing our subsequent analyses , the absence of reported features correlated with a lower number of reported publications ( S2A Fig ) . This initial result illustrates limitations in experimental approaches and a surprising degree of uncertainty that remains about human genes and the existence of their gene products . To assess whether the values of these features , rather than solely their presence , would quantitatively inform on the number of publications of individual genes , we proceeded by only considering the 12 , 948 genes with a complete set of features ( S2 Table ) . Using gradient boosting regressions with out-of-sample Monte Carlo cross-validation [25] , we could predict to a significant extent the number of publications on any given gene ( Fig 1A , Spearman: 0 . 64 ) . Remarkably , 15 out of 430 features contributed the most to our model’s accuracy ( S3A Fig ) and fell into six categories that specify the abundance of gene-encoded RNA and protein molecules across multiple tissues ( RNA abundance in adrenal glands , appendix , brain , and liver; fraction of tissues with detectable RNA expression; and protein abundance in HeLa cells ) , the positive charge of proteins , the hydrophobicity of proteins , the sensitivity of genes towards mutations ( incidence rate of missense mutations in human populations , incidence rate of loss-of-function mutations in human populations , tolerance against homozygous or recessive loss-of-function variation in human populations , CRISPR score in KBM7 cells ) , the length of the corresponding transcript and gene , and the presence of signal sequences that promote the translocation of nascent proteins into the endoplasmic reticulum . These 15 features are sufficient to account for the model’s accuracy because models using exclusively those features yields prediction accuracies highly comparable to those of the full model when trained on the same 12 , 948 genes with a complete catalog of features ( Spearman: 0 . 61 , S3B Fig ) , or on all 15 , 056 genes on which these 15 features are defined ( Spearman: 0 . 59 , S3C Fig ) . We therefore used these 15 features to define a 15-dimensional space for the 15 , 056 genes that reflects the correlation between publications and individual features and combinations of distinct features ( S3 Table ) . Clusters of genes within this space were enriched for distinct Gene Ontology annotations and thus known biological roles ( Fig 1B , S4 Fig ) . This initial finding demonstrates that the number of publications on genes can be attributed in a large extent solely to a small set of their physical , chemical , and biological characteristics . The 15 features described above have all been suspected to affect the ability to study specific genes by traditional methodologies [23 , 26–28] . Prompted by this fact and ample sociological observations on science , that the “rich” can get “richer” [9 , 29] , we next detailed the consensus between the overall number of publications per gene and past research . In line with the similarity among prior reports on the disparity in the number of publications per gene , we found that the present inequality in the number of publications has stayed constant since the year 2000 ( S5A and S5B Fig ) . Similarly , we found the number of publications per gene to be highly correlated between the current decade and preceding time periods of research ( Fig 2A , Spearman: 0 . 84 ) . Interestingly , we also identify six genes that are presently experiencing a strong increase in their number of publications , which can be traced back to a recent acknowledgment of their medical importance ( S4 Table ) . In contrast to the alternative hypothesis that research patterns on human genes would be particularly dynamic [1 , 2] , and generalizing beyond earlier studies on two gene families [6 , 21] and genes expressed specifically in the brain [30] , we find that human genes that had been reported early—as indicated by an early initial publication date on the genes or their encoded gene products [19 , 31]—tend to also be more studied presently ( S5C Fig , Spearman: 0 . 58 ) . For example , all genes that had been reported upon by 1991 ( corresponding to 16% of all genes ) account for 49% of the literature of the year 2015 ( S5D Fig ) . Initial reports further add to the predictability of the number of publications as an inclusion of their year improves the models’ accuracy ( Fig 2B , Spearman: 0 . 75 ) . To identify the factors associated with the initial reports of genes , we next created separate models with the above 430 features and trained them to predict the year of initial publications . While these predictions are slightly less accurate ( Fig 2C , Spearman: 0 . 48 ) than predictions on the number of publications , the underlying models again selected for highly similar features—most prominently , the presence of signal peptides , the abundance of transcript and protein molecules , and the sensitivity towards mutations ( S5E Fig , S5 Table ) . This shows that characteristics of genes , which have been important for the initial discovery of genes , remain partially correlated with the number of present publications on those genes . Similarly , we observe that while the number of publications is correlated between the first entry ( e . g . , AKT1 ) and the second entry ( e . g . , AKT2 ) of a gene family ( S5F Fig , Spearman: 0 . 69 ) , first entries have more publications ( Mann–Whitney U test: p-value < 10−24 ) . This demonstrates that even among evolutionary and chemically highly related genes , early initial reports coincide with a higher number of publications ( S5F Fig ) . Yet , the reduced prediction accuracy observed for the prediction of the year of the initial report may hint at the presence of another factor or factors that were not included in our curation of 430 gene-intrinsic features . Thus , we performed a bibliometric analysis of PubMed to compare individual publications against the genes contained in the publications that they cite . Focusing on the publications reporting the discovery of new human genes , we found an overrepresentation of publications that cite studies of nonhuman genes ( Figs 2D and S6A ) . Inspecting the organisms of these genes , we observed two classes of organisms . The first class preferentially co-occurred together with human genes and consisted of Mus musculus , Rattus norvegicus , Bos taurus , and Gallus gallus ( 37% , 9 . 1% , 2 . 6% , 2 . 5% of all citations , respectively ) . The second class preferentially occurred in publications without human genes and consisted of Drosophila melanogaster , Saccharomyces cerevisiae , Escherichia coli , Xenopus laevis , Caenorhabditis elegans , and Schizosaccharomyces pombe ( 22% , 10% , 4 . 0% , 2 . 5% , 1 . 6% , 1 . 5% of all citations , respectively ) ( S6B Fig ) . Assuming that citations are one proxy of scientific impact , this finding suggests that initial reports on human genes have been particularly influenced by research in model organisms and that multiple model organisms have contributed complementary roles in the discovery of human genes . With these insights , we dramatically increased the prediction accuracy of the year of initial report of a human gene by including the years of the initial reports on homologous genes of model organisms ( Fig 2E , from Spearman: 0 . 48 to 0 . 71 ) . Moreover , the years of the initial reports on homologous genes improved prediction accuracy of the number of publications to a greater extent than the year of the initial report on the human genes themselves ( S7A Fig , Spearman: 0 . 81 ) . Consistent with the picture emerging from these analyses , the homologous genes of unstudied human genes are likewise unstudied in model organisms ( S6 Table ) , and including the number of publications on homologous genes yielded almost perfect predictions of the number of publications for individual human genes ( Fig 2F , Spearman: 0 . 87 ) , while human-specific genes without homologous genes remain significantly less studied ( S7B Fig , Mann–Whitney U test: p-value < 10−32 ) . Taken together , these findings demonstrate the impact of research on model organisms on the knowledge acquired on human biology—a hypothesis that had been proposed but not demonstrated previously [32] . Given the observed historic continuity of scientific endeavors , we wondered whether biomedical research has already identified all particularly important human genes and hence allocates the production of publications accordingly . We follow the naïve assumption that researchers distribute their attention equally across all genes contained in the same publication ( S8 Fig ) . Despite this simplifying assumption , we reassuringly observe that genes that have received the most attention in publications are around three to five times more likely to be sensitive to loss-of-function mutations or to have been identified in genome-wide association studies ( GWAS ) ( Fig 3A ) . This enrichment is greatest for genes that have been repeatedly identified by several independent studies on the most frequently studied human phenotypic traits . However , we observe an extraordinarily more extreme 13-fold enrichment in the average attention when comparing the genes that have received the least attention to those genes that have received the highest attention ( Fig 3A ) . Hence , while biomedical research does focus on important genes , a disproportionally high amount of research effort concentrates on already well-studied genes . We observe a similar pattern when inspecting the allocation of funding by the National Institutes of Health ( NIH ) as another proxy of importance . Although not surprising given the correlation between the number of publications per gene and the amount of funding allocation by the NIH ( S9A and S9B Fig , Spearman: 0 . 95 ) , the above modeling strategy accurately predicts the allocation of billions of research dollars ( Fig 3B , Spearman 0 . 70 ) , and would do so particularly well for genes supported by multiple grants ( S9C Fig ) . Yet , prediction accuracy only marginally improves by additionally considering 3 , 176 features detailing known annotations between genes and diseases ( S9D Fig , Spearman: 0 . 73 ) , and is greatly—but not completely—impaired if only considering the latter ( S9E Fig , Spearman 0 . 43 ) . This shows that the previously uncovered intrinsic characteristics of genes and the year of the initial report of homologous genes not only correlate with research funding , but that they would do so to a larger extent than presently existent knowledge about the role of genes in disease . Along the same lines , if exclusively considering genes with a reported role in disease , we found that the same models that had predicted the year of the initial publication of genes ( Fig 2E ) also predicted the likelihood of the existence of both approved and preclinical drugs ( Fig 3C , S9F Fig ) . Collectively , these findings show that a small number of characteristics of genes and the availability of model organisms exert a strong influence on basic and applied research on human disease and that the resulting research can significantly deviate from the actual biological importance of individual genes . The strong correlations uncovered , and earlier work on the availability of reagents [5 , 6 , 21] suggest , that researchers may face very practical constraints that prevent them from exploring little-studied genes and that there might be a need for alternative discovery strategies [33] . In support of this possibility and extending beyond the above findings on the bulk of accrued knowledge , we observe that the fraction of genes that have been described in focused single-gene studies has only been increasing at a constant rate ( Fig 4A ) . Extrapolating from this trend , we estimate that it would take at least five decades until all genes are sufficiently studied . Similarly , simply studying little-studied genes might not be very informative and could expose junior scientists at an increased career risk ( S10A Fig ) . Along the same lines , grant categories of the NIH dedicated to exploratory research , which do not require preliminary data , and grants categories dedicated to innovative research or the training of scientists all closely reproduce the imbalance observed for the biomedical literature , with 5% of the human protein-coding genes accounting for half of the publications ( S10B and S10C Fig ) . Given a recent bibliometric study , which demonstrated that novelty could , however , be beneficial for the impact of a scientific publication if combined with an established research context [34] , we therefore thought to build a resource that provides a context for the exploration of little-studied genes . Inspecting the properties of existing publications on little-studied genes , we found that these genes tend to occur in large-scale investigations that include most genes ( S11A and S11B Fig ) . Hinting at an ability of large-scale studies to support research on less investigated genes , we observed that these studies serve as a frequent reference for other publications ( Fig 4B , S11C Fig ) and that single-gene studies that refer to them tend to focus on genes that are less studied than those genes contained in single-gene studies that refer to single-gene studies ( S11D Fig ) . To determine the extent to which large-scale collections of biological information could already serve as potential starting points for detailed characterizations on most genes , we next extended our resource with databases—such as a collection of public RNA interference ( RNAi ) experiments [35] , a catalog of human protein complexes [36] , and a catalog of public differential gene expression experiments [37]—that could potentially be affected by biased experimental choices . We find that the 27% of genes that have never been studied by a full publication ( S12A Fig ) are less frequently identified in publicly available data of large-scale experiments and that they are less likely to have characteristics associated with a high number of publications ( Fig 4C , S12B Fig ) . However , we also find that there already exist gene-specific data on possible experimentation for 83% of them and that for 25% of them , there exist at least three qualitatively distinct types of data ( S12C Fig ) . This strongly suggests that the characteristics of genes and homologous genes that prevented their early discovery would no longer prevent their more detailed study . To facilitate exploration and hypothesis generation , we provide a curated guide that specifically directs to the appropriate sources of gene-specific preliminary data ( S7 Table ) . Our analysis further shows that distinct large-scale approaches cover distinct areas of the 15-dimensional space , with genes identified in high-throughput interaction studies being strongly enriched in regions containing abundantly expressed genes [23] , and genes identified through differential expression studies being enriched in regions containing genes whose transcripts are ubiquitously detected in adult tissues through current technology . In contrast , genes identified through their phenotypes within loss-of-function RNAi screens cover the 15-dimensional space more evenly ( Fig 4D ) . Similarly , genes with a highly reproduced association to genetic traits cover multiple areas of the 15-dimensional space , some predicting a large number of publications and others predicting a small number of publications ( Fig 4E , S4 Fig ) . For illustration , consider the RNA of the heavily studied gene , TERT , the catalytic subunit of telomerase , which is undetectable in most adult tissues . While our analysis shows that this biological characteristic is generally associated with a low number of publications , the absence of TERT restricts excess cell proliferation [38]—a factor that overcomes the difficulty in its study following its ectopic activation . Another interesting illustration is provided by the poorly studied breast cancer gene CCDC170 , which encodes for one of the most charged and acidic human proteins but also appears to have some structural role in maintaining the organization of Golgi-associated microtubules [39] . As a final illustration , consider C1orf106 , a gene with the second-strongest genetic association to ulcerative colitis . Despite being among the top 20% of genes with the most frequently identified associations in differential gene expression experiments ( S7 Table ) , C1orf106 had never been followed up until recently , when gene-specific pull-down experiments revealed its role in the regulation of the stability of epithelial adherens junctions [40] . This demonstrates that functional studies remain a powerful strategy to discover novel biology that does not reproduce past research biases . To provide a broader perspective on the strategic options for further exploration , we next introduced aggregate measures for the presence of genetic support and experimental approachability and the existence of homologous genes in invertebrate model organisms . While some of the initially identified clusters ( Fig 1B ) seem experimentally well accessible in humans or model organisms , other clusters seem resilient to those approaches ( Fig 4E ) . An opposite example is a cluster enriching for transcriptional coactivator activity . It contains several evolutionarily conserved genes that are highly sensitive towards loss-of-function mutations and experimentally approachable . This cluster contains multiple highly studied modulators of cellular physiology , such as the genes MTOR , CLTC , TAF1 , and CREBBP . However , this cluster also contains DICER1 , which catalyzes the maturation of microRNAs and is a recent recipient of research attention , and whose discovery was perceived as an enormous surprise following a long-held lack of attention towards non–protein-mediated gene regulation [41] . Intriguingly , this cluster includes two still mostly uncharacterized members of large gene families , IPO9 and ANKRD52 . This lack of attention illustrates that even genes with seemingly promising characteristics can remain mostly ignored . To facilitate identification of such genes , we are also providing a list of these genes ( S8 Table ) and a map that identifies them within the vicinity of custom sets of genes ( S9 Table ) . We further add another map that allows probing custom sets of genes for the above aggregate measures ( S10 Table ) .
Because the difficulty of pursuing different research directions varies both within distinct fields of biological and nonbiological inquiry [16] , we suspect that our findings may be generalizable to other areas of science . For example , mathematics dealt for centuries nearly exclusively with “smooth” curves; only in the last half century did it address the study of infinitely rough curves [42] . Our work demonstrates that even highly promising genes that could already be studied by current technologies remain ignored . This suggests that the ossification of past research topics [43] , which for human genes becomes apparent at the turn of the millennium ( S5A and S5B Fig ) , reflects upon processes that extend beyond past experimental possibilities . Indeed , a recent seminal bibliometric study on 250 scientific fields , including molecular biology , demonstrated that scientific fields move from a phase characterized by “the rich get richer” towards a phase of ossification as the annual number of publications increases [43] . Our study provides empirical support for the presence of several processes that could possibly contribute to this ossification , including but not limited to the availability of prior knowledge [7]; biases in computational annotations; the availability of reagents [6 , 21]; the career prospects of junior researchers; the support by grants [3]; training agendas; the presence of an overwhelming set of competing future research options [43]; a slow transition of research between large-scale studies and small-scale studies [44 , 45]; a sustained ease to experimentally study certain genes; a shortage of large-scale studies that attribute function through perturbing genes and monitoring altered physiology rather than through guilt by association [46 , 47]; and a decrease in the workforce that uses model organisms , which accelerated around the year 2000 in favor of an increased fraction of scientists that exclusively work on human genes ( S13 Fig ) . Similarly , our work shows that , with some rare exceptions , the human genome project did not suffice to promote an exploration of novel genes and the biology encoded by them . Given their presence in the human genome , it is certain that the majority of protein-coding genes have biological relevance [48] . For some genes the relevance might be apparent , such as for the δ- and β-globins [49 , 50] , which mark among the first human genomic clones and encode for the hemoglobin subunits . For other genes , most of their physiological relevance might only unfold after their basic characterization outside of medical contexts , such as for the heat shock-inducible gene HSP70 , which marks an important subsequent human genomic cloning endeavor [51] and participates in a network of genes that control protein homeostasis—a process whose failure characterizes aging in humans and model organisms , and a basis for diseases of protein conformation [52] . Furthermore , many current insights on biology relate to monogenic experimentation schemes , whereas biological processes appear polygenic , which could plausibly further contribute to the continued inability to explain many of the biological processes known to occur [53] . Indeed , our work supports the hypotheses that an insufficient understanding of the biology of many disease genes has prevented the successful development of therapies [7 , 54 , 55] and that preclinical research is biased towards experimentally well-accessible genes [28] . To visualize potentially implicit biases underlying distinct research projects and findings , we provide a copy of the 15-dimensional feature space , whose regions correspond to distinct biases ( S4 Fig , S3 Table ) . In order to accelerate the pace of discovery , we propose the need for funding mechanisms of scientists and calls for proposals that encourage the pursuit of nonredundant and likely highly unpredictable research directions . In order to counter the career forces currently pushing towards conformity , there would be a need for stable , long-term support for such innovators to focus on the unknown . Just as the Royal Society sponsored target studies of the unknown with an eye towards the economic potential of certain discoveries , we also predict that exploring the uncharted territories of unknown biology by investigating unstudied and understudied genes will yield satisfying observations that would contribute economically and medically . We believe that the resource presented here provides a jumping point for further systems-level investigation on the formation of scientific knowledge [56] and a guide to researchers who want to identify promising but little-studied genes .
Linkage of genes to publications was obtained from NCBI NIH ( https://ftp . ncbi . nlm . nih . gov/gene/DATA/gene2pubmed . gz ) in early 2017 . Patent data were obtained from Rosenfeld and Mason [57] . Gene Ontologies , mapped to Entrez Gene IDs , were obtained from NCBI in early 2017 ( https://ftp . ncbi . nlm . nih . gov/gene/DATA/gene2go . gz ) . Funding information was obtained from NIH ExPORTER ( https://exporter . nih . gov/ ) in early 2017 . Names of genes and chromosomes were obtained from NCBI NIH in early 2017 ( https://ftp . ncbi . nlm . nih . gov/gene/DATA/gene_info . gz ) . Article types and publication titles were obtained from MEDLINE ( https://www . nlm . nih . gov/databases/download/pubmed_medline . html ) through a local copy of their database in early 2017 . Disambiguated authorship information was obtained from Clarivate Analytics . SwissProt and TrEMBL protein sequences , and mapping tables to Entrez GeneIDs , were obtained from Uniprot in early 2017 ( ftp://ftp . uniprot . org/pub/databases/uniprot/current_release/knowledgebase/complete/uniprot_sprot . fasta . gz , ftp://ftp . uniprot . org/pub/databases/uniprot/current_release/knowledgebase/complete/uniprot_trembl . fasta . gz , ftp://ftp . uniprot . org/pub/databases/uniprot/current_release/knowledgebase/idmapping/idmapping_selected . tab . gz ) . Linkage tables between Entrez Gene IDs and Ensembl Gene IDs were obtained from NCBI NIH in early 2017 ( https://ftp . ncbi . nlm . nih . gov/gene/DATA/gene2ensembl . gz ) . Genes , coding sequences from genomic RNA , and validated RNA sequences were obtained from Genbank ( Genome version GRCh38 . p10 ) ( ftp://ftp . ncbi . nlm . nih . gov/genomes/all/GCF/000/001/405/GCF_000001405 . 36_GRCh38 . p10 ) using a manually reviewed definition of reference chromosomes according to https://ncbi . nlm . nih . gov/genome . Allele frequencies in human populations were obtained from the ExAc database [58] . Compartment information and protein abundance were obtained from Itzhak and colleagues [59] . Loss-of-function information in human cell lines was obtained from Blomen and colleagues [60] , Hart and colleagues [61] , and Wang and colleagues [62] . Thermal stability on proteins was obtained from Leuenberger and colleagues [63] . Transcript abundance in cells and tissues was obtained from the human protein atlas [64] . Transcript stability was obtained from Tani and colleagues [65] . GWAS were obtained from the NHGRI-EBI Catalog v1 . 0 [37] . A local copy of the Web of Science Database was obtained from Clarivate Analytics ( and formerly Thomson Reuters ) . Homologene Version 68 was obtained from NCBI NIH ( https://ftp . ncbi . nlm . nih . gov/pub/HomoloGene ) . Associations between genes and diseases were obtained from Genecard’s GeneALaCart service ( https://genealacart . genecards . org ) in early 2017 through successive batch queries with all official human ( HUGO ) gene symbols . The BioGRID database [66] was obtained from BioGRID ( Version BIOGRID-3 . 4 . 147 ) . Drugs and their targets were obtained from DrugBank ( Version 5 . 0 . 7 ) . Bioplex 2 . 0 complexes were obtained from Huttlin and colleagues [36] . GenomeRNAi v17 was obtained from www . genomernai . org . EBI Gene Expression Atlas ( GXA ) was downloaded in spring 2017 from www . ebi . ac . uk/gxa . For genes , we determined the fraction corresponding to every nucleobase , and the combined ratio of cytosine and guanine , and counted the number of all nucleobases . For protein-coding sequences , we additionally determined the fraction corresponding to individual codons and measured the codon bias according to multiple methods [67–70] . For transcripts , we obtained FPKM values from Uhlen and colleagues [64] and additionally determined the fraction of samples with an expression below 1 FPKM analogously as a surrogate for detectable expression [64] . For SwissProt and TrEMBL proteins , we determined the fraction of the primary sequence covered by individual amino acids . Moreover , we used BioPython [71] to determine the fraction of acidic , aromatic , basic , charged , helix affine , hydrophobic , polar , uncharged polar , sheet affine , and turn affine amino acids . We further used BioPython to estimate protein GRAVY , the protein’s isoelectric point , and molecular weight . Additionally , we counted the total amount of amino acids and thus the length of the protein . We used the Python version of RADAR [72] with its default settings to measure the total number of repeats , and the total RADAR score , and the length of the highest scoring repeat . We used SEG [73] ( from NCBI’s ftp . ncbi . nlm . nih/pub/ ) with its default settings to measure the total amount of amino acids , the fraction of the protein residing in low complexity regions , the length of the longest low complexity region , and the fraction of the protein covered by the longest low complexity region , and counted the total number of low complexity regions and the number of low complexity regions longer than 5 , 10 , 20 , and 40 amino acids . We used SignalP [74] with its default settings to determine the presence of a predicted cleavage site , the maximal cleavage score , the presence of at least four transmembrane residues , and the nucleotide position of the mature protein . In the absence of measurements on transcript expression and stability , we used −1 to indicate the presence of a low expression . In the absence of a SwissProt protein entry for a gene , TrEMBL protein entries were used for a given gene . In the absence of measurements on protein localization and stability and protein abundance , we used −1 to indicate the presence of a low expression . Information of genes and gene products was mapped to Entrez GeneIDs . Only unambiguous mappings were considered . In the case of multiple entries mapping to a single Entrez GeneID ( e . g . , multiple transcripts encoded by the same gene ) , we used the median of the features . Unless specified otherwise ( for reviews ) , we considered publications that were ( a ) assigned by MEDLINE to correspond to a “case report , ” “classical article , ” “clinical trial , ” “clinical trial phase I , ” “clinical trial phase II , ” “clinical trial phase III , ” “clinical trial phase IV , ” “comparative study , ” “historical article , ” “journal article , ” “meta analysis , ” “multicenter study , ” “randomized controlled trial , ” “twin study , ” or “validation study”; ( b ) were further not assigned by MEDLINE to also be a “review”; and ( c ) were further not occurring in a journal in which 50% or more of all articles were assigned by MEDLINE to be a “review . ” We considered protein-coding genes of Homo sapiens ( NCBI taxonomy ID: 9606 ) that would also contain an official HUGO symbol and be featured in at least one reference research publication . Features were z-scored across the genes and clustered using Ward’s method . We predicted the log10-transformed number of publications and z-scored the features across genes . We used 90% of the genes as training data and predicted the remaining 10% . We performed at least 400 randomizations using randomly chosen subsets without replacement . This corresponds to a number of iterations in which , within initial test runs , we would not observe changes in the pooled readout within the number of digits provided in this publication . We used Scikit-learn’s [25] ( version 0 . 19 ) Gradient Boosting Regressor with 300 estimators and a Huber loss function . The results of individual randomizations for individual genes were pooled by taking the median . We considered the 15 features with the highest median importance to the gradient boosting regression models . We considered all reference genes for which these 15 features were defined and z-scored every feature separately across these genes . Grouping onto two dimensions was done by Scikit-learn’s implementation of the t-distributed stochastic neighbor embedding [75] . We considered entries to be negating if the qualifier started with NOT , or if the evidence code was “ND . ” For temporarily valid , computationally predicted entries , we considered the “IEA” and “RCA” evidence codes . We excluded unmapped entries with the evidence code “–” or “NR . ” Highlighted groups were chosen manually to reflect areas with higher local concentration . Terms considered for enrichment were non-negating , non-temporary Gene Ontology annotations with mapped evidence . We used an EASE score [76] , an observation-corrected variant of Fisher’s exact test , and determined the false discovery rate through Scikit-learn’s implementation of Benjamini and Hochberg’s procedure using an alpha of 0 . 05 [25] . To account for an uneven total number of pairs between genes and publications , when defining the enrichment within recent years , we normalized either time interval to have the same number of pairs between genes and publications . We performed a manual literature review on genes with the highest log2 fold change in the number of publications , upon filtering for the presence of at least 10 publications in the interval between 2011 and 2015 . We performed a manual literature review and citation analysis to identify findings that changed research on those genes in the subsequent years . Genes highlighted in the main figure were chosen manually to cover a broad range of different numbers of publications , while a complete list is given in S4 Table . The prediction of the year itself was done as described above for the prediction of the number of publications . When adding discoveries of homologous genes , we considered the years of the first description of homologous genes and the years of the first single-gene publications of homologous genes of model organisms listed in Homologene , and indicated absent values ( indicative of the absence of either a homologous gene or publications ) by assigning the value −1 . Confidence intervals of 95% reflect bootstrapped estimates as computed by Python’s seaborn package [25] ( versions 0 . 7 and 0 . 8 ) . We defined publications with a discovery of a new human gene as those publications that would report on a gene within the year in which the first report on the same gene would appear . We counted the number of cited publications that would have at least one human gene , and the number of cited publications that would have at least one nonhuman gene . For analyses showing the fraction of literature , we performed a fractional counting of publications . Rather than counting every publication as 1 towards every gene , the value of a publication towards a given gene would be 1/ ( number of genes considered in the publication ) . We considered genes that would not map to a Homologene group with at least one nonhuman gene . The analysis only included genes with a human Entrez Gene ID that would be smaller or equal to the highest human Entrez Gene ID within the Homologene Database and thus could have been considered for Homologene . We performed a fractional counting of publications . Enrichment was calculated as the log2 fold change over the ( fractional count of publications in indicated time frame ) / [ ( total number of publications in indicated time frame ) / ( number of reference genes ) ] . We considered EBI’s mapping of associations and only considered associations lying within the sequence of one , but not multiple , genes . We counted the occurrence of at least one association per publication between a gene and a trait . For strong association , we only considered traits covered in at least 10 distinct studies and genes that would be associated with more than 20% of the studies for such a trait . We considered genes with a pLI over 0 . 9—a threshold that the authors [58] describe as “extreme loss-of-function” intolerance on their accompanying web portal . We considered NIH funding information between 1985 , the year in which data of grants would be provided at the resolution of principal investigators , and 2015 . We performed inflation correction using the average United States consumer price index . We equally distributed the total money allocated to a given NIH project ID to all publications supported by this project , and subsequently within the individual genes in this project . We used disease associations from Malacards for Unified Diseases , Orphanet , Human phenotypes , and OMIM as disease linkage features and constructed additional features that would count for the total number of entries within each of the four data sets . Because of computational constraints , we subsequently removed disease linkage features with fewer than 10 genes . Notably , prediction accuracy did not improve if keeping all linkages of Unified Diseases ( Spearman 0 . 73 for addition on top of other features—analogously to S9D Fig; Spearman 0 . 42 for exclusive usage—analogously to S9E Fig ) or OMIM ( Spearman 0 . 71 for addition on top of other features—analogously to S9D Fig; Spearman 0 . 16 for exclusive usage—analogously to S9E Fig ) . As the rank of the popularity , we used the fractional count of publications up until the indicated year . We only considered publications of authors that have not yet transitioned to a principal investigator status . As principal investigator status , we consider authors that have at least two last author publications with at least one fellow coauthor . We matched publications contained in MEDLINE to records from Web of Science in a two-step process: ( a ) if available , we used the digital object identifier ( DOI ) , allowing for an unambiguous identification of publication entries; ( b ) otherwise , given the MEDLINE record , we retrieve all publications from Web of Science with the same list of authors’ last names , and that were published in the same year and journal . We then identify the best-matching record by calculating the Levenshtein distance ( implemented in seatgeek’s FuzzyWuzzy Python package: https://github . com/seatgeek/fuzzywuzzy ) between titles of the MEDLINE and the Web of Science record , respectively . We only considered publications that would map unambiguously and had a mapping score of at least 95 ( maximum score 100 ) . In total , for 97% of all publications in MEDLINE containing a reference to a gene , we were able to identify the corresponding record in Web of Science . Following Uzzi and colleagues [34] , we counted citations over the 8 years following the year of the publication . Publications with more than two authors and publications with consortium as the sole affiliation were considered to be team publications . For the analysis of BioGRID , we considered BioGRID entries that had been associated with at least one gene in MEDLINE and counted the unique genes of a publication—after pooling the indicated gene A and gene B entries of an interaction—which would usually be indicative of bait and hit , respectively . Western blots following affinity purification were obtained from BioGRID . For differential gene expression analysis , we used EBI GXA and considered genes to be differentially expressed if their ( nonadjusted ) p-value would be below 0 . 0001 . For RNAi , we only considered phenotypes that were not measured through distinct shRNA abundance and only considered genes occurring in at least 20 studies ( which could possibly have monitored distinct phenotypes ) . We considered a gene to have a strong RNAi if more than 30% of the studies containing the gene would report a phenotype for this gene . This was motivated by the ( not shown ) observation that genes fall into a bimodal distribution according to the fraction of studies reporting a phenotype , separated at the chosen threshold of 30% . Code for the curation of data sets and for analysis is available at github . com/tstoeger/plos_biology_2018_ignored_genes . | Biomedical research is one of the largest areas of present-day science and embeds the hope and potential to improve the lives of the general public . In order to understand how individual scientists choose individual research questions , we study why certain genes are well studied but others are not . While it has been previously observed that most research on human genes only concentrates on approximately 2 , 000 of the 19 , 000 genes of the human genome , the reasons for this ignorance are largely unknown . We systematically test explanations for this observation by compiling an extensive resource that characterizes biomedical research , including but not limited to hundreds of chemical and biological properties of gene-encoded proteins , and the published scientific literature on individual genes . Using machine learning methods , we can predict the number of publications on individual genes , the year of the first publication about them , the extent of funding by the National Institutes of Health , and the existence of related medical drugs . We find that biomedical research is primarily guided by a handful of generic chemical and biological characteristics of genes , which facilitated experimentation during the 1980s and 1990s , rather than the physiological importance of individual genes or their relevance to human disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"genome-wide",
"association",
"studies",
"rna",
"interference",
"citation",
"analysis",
"model",
"organisms",
"scientists",
"experimental",
"organism",
"systems",
"genome",
"analysis",
"epigenetics",
"molecular",
"biology",
"techniques",
"science",
"and",
"technology",
"w... | 2018 | Large-scale investigation of the reasons why potentially important genes are ignored |
The p53 tumor suppressor is a sequence-specific pleiotropic transcription factor that coordinates cellular responses to DNA damage and stress , initiating cell-cycle arrest or triggering apoptosis . Although the human p53 binding site sequence ( or response element [RE] ) is well characterized , some genes have consensus-poor REs that are nevertheless both necessary and sufficient for transactivation by p53 . Identification of new functional gene regulatory elements under these conditions is problematic , and evolutionary conservation is often employed . We evaluated the comparative genomics approach for assessing evolutionary conservation of putative binding sites by examining conservation of 83 experimentally validated human p53 REs against mouse , rat , rabbit , and dog genomes and detected pronounced conservation differences among p53 REs and p53-regulated pathways . Bona fide NRF2 ( nuclear factor [erythroid-derived 2]-like 2 nuclear factor ) and NFκB ( nuclear factor of kappa light chain gene enhancer in B cells ) binding sites , which direct oxidative stress and innate immunity responses , were used as controls , and both exhibited high interspecific conservation . Surprisingly , the average p53 RE was not significantly more conserved than background genomic sequence , and p53 REs in apoptosis genes as a group showed very little conservation . The common bioinformatics practice of filtering RE predictions by 80% rodent sequence identity would not only give a false positive rate of ∼19% , but miss up to 57% of true p53 REs . Examination of interspecific DNA base substitutions as a function of position in the p53 consensus sequence reveals an unexpected excess of diversity in apoptosis-regulating REs versus cell-cycle controlling REs ( rodent comparisons: p < 1 . 0 e−12 ) . While some p53 REs show relatively high levels of conservation , REs in many genes such as BAX , FAS , PCNA , CASP6 , SIVA1 , and P53AIP1 show little if any homology to rodent sequences . This difference suggests that among mammalian species , evolutionary conservation differs among p53 REs , with some having ancient ancestry and others of more recent origin . Overall our results reveal divergent evolutionary pressure among the binding targets of p53 and emphasize that comparative genomics methods must be used judiciously and tailored to the evolutionary history of the targeted functional regulatory regions .
Since the completion of the human genome , cataloging transcription factor binding sites ( TFBSs ) has been critical for understanding gene regulation . The use of comparative genomics ( evolutionary conservation across species ) is often championed as a method to separate the functional regulatory sequence “wheat” from the nonfunctional “chaff” [1] . As the number of mammalian full genome drafts increases , the integration of TFBS predictions with lists of conserved noncoding regions ( CNCs ) has emerged as a key step in the TFBS identification process [2–4] . If TFBS predictions are contiguous to DNA features that coincidentally have a critical structural role such as maintenance of chromatin organization , the appearance of conservation may be intensified even further . Although these methods have greatly enhanced our knowledge of the human genome's regulatory repertoire , overreliance on conservation information can potentially exclude genuine binding sites . Since TFBSs are typically small , they can arise by chance in a gene's promoter and therefore may decrease selective pressures to maintain already existing sites [5] . Another wrinkle is that the evolutionary forces that created the conservation blocks may no longer be functionally relevant to humans [6] . Additionally , recent scans for natural selection in human gene coding regions have revealed that distinct biological pathways often are subject to widely different evolutionary pressures [7 , 8] , particularly since mutation rates have been shown to vary across the genome [9] . Genes involved in oncogenesis and tumor suppression have experienced recent selection for mutation in primate lineages [7 , 8 , 10] . DNA binding sites of transcription factors are also functional components of these pathways and are likely under similar evolutionary pressures . Indeed , we have focused recently on identifying human single nucleotide polymorphisms that alter the function of transcription factors [11 , 12] . As a result , we have investigated the assumptions for using mammalian conservation as an obligatory screening step for seeking TFBSs . The p53 tumor suppressor gene , encoded by the TP53 master regulatory gene , is a transcription factor that coordinates a network of cellular responses to environmental insults . Over half of human cancers have a mutation in the p53 protein or one of its partners [13] . The p53 protein is estimated to have several hundred transregulation target genes that affect pathways including apoptosis , DNA damage repair , and cell-growth arrest [14] . As a result , p53 target genes are highly sought-after drug targets for halting cancer progression . According to in vitro experiments , the p53 protein binds specifically to a palindromic consensus sequence , RRRCWWGYYY ( N0−13 ) RRRCWWGYYY [15] , with nearly all REs containing at least one mismatch; in vivo results have suggested that the spacer region may be much smaller [14 , 15] . The sequence is typically located within 5 , 000 bases of the target gene's transcriptional start site , and p53 either induces or represses expression upon p53 binding [16 , 17] . One feature of p53 that confounds the discovery of novel transregulated genes is that while some binding sites match the expected consensus sequence quite well , others can be consensus poor and yet are both necessary , and sufficient , to transactivate a gene [18] . A recent study has suggested that the “rules of engagement” for p53 REs may differ based on the activated pathway , particularly in the apoptosis and cell-cycle–related systems [19] . Thus , we have used cross-species conservation to examine if these groups of elements exhibit distinct conservation profiles . To evaluate the utility of comparative genomics approaches in the identification of potential p53 target REs , we gleaned the literature for a high quality set of bona fide p53 REs to estimate the degree of conservation between humans and other mammals . To relate the TP53 system to other master regulators , we compare its binding site conservation to those of the transcription factors encoded by two other genes: NFκB ( nuclear factor of kappa light chain gene enhancer in B-cells ) , central to inflammation responses , and NFE2L2 , which encodes NRF2 ( nuclear factor [erythroid-derived 2]-like 2 nuclear factor ) , a regulator of oxidative stress . Their repertoire of interactions is expected to be highly preserved throughout the mammalian lineage . The NFκB transcription factor is a heavily studied biological switch of the inflammation , apoptosis , and immune responses [20 , 21] . It binds the consensus sequence GGGRNNYYCC [22 , 23] , and its signaling system is highly conserved even when examined in invertebrates [21 , 24] . NRF2 binds antioxidant REs ( consensus sequence = TMANNRTGAYNNNGCRWWWW [25] ) that are comparable in size to those of p53 , show high levels of conservation [26] , and are found in the promoters of genes that confer protection from oxidative stress and chemical carcinogens [27] . Mouse models of Nrf2-dependent response to oxidative and electrophilic insults have been used to study function [28 , 29] . Additionally , the Nrf2 pathway in zebrafish operates similarly to humans and underscores the likelihood of high conservation in regulatory binding sites [30] . Because the NFκB and NRF2 binding sites were determined to be highly conserved , these two sets of TFBSs serve as positive controls in estimates of conservation . Our comparative genome analysis , which includes a coincident evaluation of sampled promoter sequences and coding region sequence , reveals that mammalian conservation does not apply to p53 target REs in general . However , among subgroups of target genes we observe purifying selection acting on a number of p53 binding sites , including many cell-cycle–related genes , while rodent to human homology is lacking for p53 REs in apoptosis-related genes .
The literature was scanned for validated p53 ( 83 , Table 1 ) , NFκB ( 21 , Table 2 ) , and NRF2 ( 21 , Table 3 ) binding sites associated with human genes . Human genome coordinates were located and then referenced within global multiple alignments held at the University of California Santa Cruz [UCSC] ( California , United States ) genome browser website [31] to find their corresponding locations in eleven other mammals . Using these global alignments , percent sequence identity was calculated for each of the 125 binding sites across the eleven mammals , with the calculation adapted to reflect consensus sequence degeneracy , since every model RE had positions where one or more of the four nucleotides could be tolerated . Also , the p53 RE was unique in that the spacer region between the two half sites could be any size or sequence up to thirteen bases . We therefore calculated sequence identity by omitting the p53 spacer region and ignoring mismatches in the alignments that still fit the TFBS consensus ( CDKN1A and PCNA examples shown in Figure 1 , others in Figure S3 ) . Figure 2 plots the conservation distribution for each set of human TFBSs across mouse , rabbit , rat , and dog . Although comparative data relating the chimpanzee and rhesus monkey were also analyzed , we observed , as expected , that these species were too evolutionary close to humans to be informative ( Figure S1 ) . For example , nearly any human sequence was in excess of 95% conserved in these two primates . Also graphed in Figure 2 are the results for sets of DNA sequence fragments randomly chosen from promoter ( gray ) and protein-coding ( blue ) regions . This allows TFBS conservation levels to be viewed in context of the evolutionary pressures exerted on other genomic sequences . The poorly conserved element shown in Figure 1 ( PCNA ) would fall in a lower percentage bin ( i . e . , be on the left side of the graph ) , as is the case for the randomly chosen promoter sequences . The promoter fragments are representative of the background genome sequence in which most TFBSs reside . On the other hand , if TFBSs were very well conserved , then the distribution would be right-shifted , as is the case for the protein-coding region fragments ( blue ) . To use conservation as a metric to separate true binding sites from the rest of the genome , their conservation should be significantly greater than that of randomly chosen promoter regions . The spike of TFBSs at 0%–9% identity in each panel represents species-specific sites that are essentially not present in the other mammals . For each of the human-to-mammal comparisons in Figure 2 , NFκB and NRF2 sites produced identity distributions that appear very similar to distributions from the coding region fragment group ( many sites with 90%–100% identity ) , which was representative of genome sequence under high purifying selection . This strongly suggests that NFκB and NRF2 sites may be under purifying selection . The human-to-mammal p53 site comparisons , on the other hand , produced conservation profiles in each species that have a high frequency of sites at zero percent identity and fewer with 90%–100% identity . This distribution is similar to the distribution obtained from randomly sampled promoter fragments ( gray ) , which we used to represent genome sequence under neutral selection . In mouse the p53 RE identity distribution was correlated with the promoter fragment identity distribution while NFκB and NRF2 showed less correlation with the promoter distribution ( Table S3 ) . Since p53 sites as a group were observed to have as many interspecific substitutions as the background genomic sequence , use of conservation level to predicting bona fide sites would not be effective . However , this result could be due to the fact that the set of 83 p53 TFBSs actually represents two or more subsets of p53 REs with distinct conservation profiles . A recent study hypothesized that the sequence requirements of p53 REs may differ based on the activated pathway such as apoptosis , DNA repair , cell-cycle checkpoints , or cell-growth arrest [19] . We therefore investigated if low and high percent identity values would apportion with p53 REs grouped by function , thereby detecting evolutionary divergence between p53 pathways . Among these 83 p53 REs , we carried out analysis of the two largest subgroups ( Table 1 ) , apoptosis-related ( n = 29 ) and cell-cycle/cell-growth–related ( n = 23 ) , on the basis of observations of Qian et . al [19] . Average percent identity to a consensus sequence was calculated for each of the transcription factor groups , including the p53 subsets , and compared via a two-tailed t-test assuming unequal variances between the datasets . The results are displayed as odds ratios ( OR ) in Table 4 , and OR values represent the odds that one type of human TFBS ( columns ) will be found as more conserved than a second TFBS type ( rows ) in comparison with other mammalian species . Ratios less than 1 ( e . g . , p53 apoptosis compared to all p53 sites ) suggest lower conservation of the TFBS in the row . Among all species , the relative conservation levels of NFκB compared to NRF2 sites were similar and the p-values for difference were not significant . NFκB sites were significantly more conserved than the entire set of p53 sites in mouse and dog , while NRF2 sites were significantly more conserved than all p53 sites in mouse and rat . The magnitude and statistical significance of the differences between sequence motifs in Table 4 was greatest when comparing either NRF2 or NFκB and the apoptotic p53 sites . For example , in each species , the ORs for relative conservation between either NRF2 or NFκB and apoptosis genes were all high ( OR > 3 . 0 ) and at a significance of at least p < 0 . 019 . On the other hand , the mean conservation level of the cell-cycle–regulating p53 REs were not statistically different from the NRF2 or NFκB sites . These observations imply that the mean p53 conservation level for all elements is really a combination of the effect of the two ontological subgroups . It has been proposed that an alternative method for evaluating a DNA fragment's conservation level is to ask whether it sits within a block of conservation [6] . All bona fide TFBSs examined in this study were matched against the “most conserved” track of the UCSC genome browser to ask whether a significant proportion fell within conserved blocks . Only 19 . 2% of all p53 REs mapped to these regions , while 52 . 8% of NRF2 and 57 . 1% of NFκB TFBSs could be colocated ( Table 5 ) . We also assessed the conservation of the randomly chosen promoter sequences according to this block method and used a two-tailed binomial test to calculate statistical significance . Intriguingly , all TFBS groups mapped to more blocks than the random promoter sequences except for the apoptosis subgroup of p53 REs . These data mirrored the percent identity conservation metric and again underscored that the apoptotic-regulating p53 TFBSs may not have been under purifying selection throughout mammalian evolution . Inspection of the individual alignments between human p53 REs and mouse reveal that 38% ( 11/29 ) apoptosis elements and 9% ( 2/21 ) of cell-cycle element could not be aligned with the multiz global alignment tool and thus showed zero identity . Thus for human p53 REs in genes such as BAX , FAS , PCNA , CASP6 , SIVA1 , and P53AIP1 we observed little , if any , similarity with rodent sequences , and these nonaligned sequences ( zero identity ) strongly impact the calculations we have made . Although it was informative to know how well human p53 TFBSs are conserved relative to other regulatory motifs , an aim of this study was to probe the utility of comparative genomics for authenticating binding sites predicted by computational methods . Receiver operator characteristic ( ROC ) curves were employed to demonstrate the sensitivity of TFBS prediction when qualified by conservation information . ROC curves are traditionally used to measure the quality of a binary classification algorithm , as a discrimination threshold is varied . Area under a ROC curve provides a visual representation of how well the conservation metric can classify the sets of bona fide TFBSs as true positives . For example , the area can be interpreted as the probability that when both a bona fide TFBS and a random promoter sequence of equal lengths are chosen at random , the decision function ( conservation in a species ) assigns a higher value to the bona fide TFBS . A perfect decision function would generate a curve with an area of 1 , meaning that 100% sensitivity was obtained ( i . e . , all true positives were found ) , and 100% specificity was reached ( i . e . , no false positives were generated ) . If conservation predicted TFBS authenticity no better than random chance , a line at 45 ° to the x-axis would be generated that bisects the ROC space ( area under the curve [AUC] = 0 . 5 ) , because as the threshold is raised , equal numbers of true and false positives compose the chosen set of TFBSs . A ROC curve that fell below this diagonal would indicate that conservation consistently predicted poorly , meaning that one should employ the lack of conservation as a decision classifier to authenticate TFBSs . When TFBS conservation was evaluated as a TFBS predictor in each of the four mammals , bona fide NRF2 and NFκB sites were consistently well predicted , whereas the ROC curve describing all p53 sites approached the random diagonal ( Figure 3A–3D ) . The latter implies that conservation analysis in these model organisms cannot enhance p53 binding site discovery , for the predictive capacity is only slightly better than random . For example , if a cutoff of 80% identity to mouse was employed as the rule for choosing p53 binding sites , only 43% of real p53 REs would be found , and 19% of the selections would be false positives ( Figure 3A ) . We were concerned that ascertainment bias ( e . g . , the presence of spurious REs ) in this large set of p53 sites might affect our findings . However , the predictivity level for any of these species does not change appreciably even when the p53 RE list is restricted to only the 30 best-characterized sites ( Figure S2 ) . In contrast , for NFκB an 80% mouse conservation threshold would allow discovery of 86% of real NFκB sites with a 25% false positive rate ( Figure 3A ) . As a result , human/mouse multiple sequence alignments are highly useful for identifying novel NFκB sites but not so for p53 motifs . This conclusion is reiterated in all four model organisms by statistical evaluation of the AUC calculations ( Table 6 ) . p53 curves had smaller AUCs compared to those for NFκB and NRF2 . A recent study noted that although the spacer region between half sites for p53 REs can be zero to13 bases , small spacers were overwhelmingly preferred in a distribution of spacer length derived from genome-wide chromatin immunoprecipitation experiments [14] . This suggests that REs with large spacers might not be valid , and we hypothesized that if ROC curve analysis was restricted to only p53 REs with small spacers ( presumed higher quality ) , perhaps much greater conservation would be observed . We examined the set of p53 REs having two or fewer spacer nucleotides between half sites and observed no increase in conservation ( Figure S2 ) . Not only was there no improvement in TFBS prediction sensitivity for this subset , but the ROC areas were visibly greater for rat and rabbit comparisons in the inverse set of REs ( i . e . , 3+ spacer bases ) . Thus , as judged by conservation , p53 RE spacer region length could not be considered a measure of RE quality . We then examined p53 RE conservation in light of gene ontology . When p53 REs were subdivided based on functional class , the sensitivity of interspecific conservation to predict cell-cycle/cell-growth sites improved considerably , approaching that for the NFκB and NRF2 targeted genes ( Figure 3 ) . The p53 apoptotic REs ( Figure 3 , dashed line ) , on the other hand , showed a dramatically different conservation profile . In the case of the mouse ( AUC = 0 . 616 ) and rat ( AUC = 0 . 568 ) , the ROC curve hovered just above the random line , which indicates a lack of sensitivity . For two species ( rabbit , AUC = 0 . 469 and dog , AUC = 0 . 437 ) , the sequence identity metric had an apoptosis RE discovery rate worse than random prediction . This suggests that the functional , apoptotic p53 binding sites are less conserved than randomly sampled sequences in gene promoters . This phenomenon was also observed when ROC curve analysis was carried out in other distant mammalian species ( tenrec , armadillo , elephant , and opossum ) ( unpublished data ) . One explanation for this provocative result could be that apoptotic p53 sites might actually display a slightly different consensus p53 binding site than that reported in the literature . Perhaps a better-fitting consensus would improve conservation . We aligned all p53 sites ( 83 ) , apoptotic ( 29 ) , and cell-cycle ( 23 ) p53 sites ( Figure S3 ) and generated sequence logos [32] ( Figure S3A–3C ) to identify improved patterns , but while there are small differences , none fit better than the existing consensus of RRRCWWGYYYN0−13RRRCWWGYY . Likewise , simply permitting any nucleotide at the least compositionally biased positions in this p53 RE subset ( bases 2 , 8 , 10 , and 11 of the p53 consensus ) did not improve the area under the ROC curve ( Figure S2 ) . These data emphasize not only that conservation analysis cannot improve identification of certain TFBSs like p53 , but also that subclasses of the same binding site may reflect distinct evolutionary profiles . A second approach was used to detect if conservation differed among nucleotide positions within the binding site . That is , could we observe heightened human-to-mammal interspecific substitutions or “sequence diversity” at particular locations within each TFBS consensus sequence ? To accomplish this , we aligned all TFBSs within each group ( p53 , NRF2 , or NFκB ) and calculated the positional sequence diversity , which was the percentage of aligned bases at each position that varied from the accepted consensus sequence ( Figure 4 ) . For example , in Figure 4A , the first position in the p53 consensus sequence differed from a purine base ( R ) at the equivalent position in the mouse in 35% of all p53 TFBSs , while positional diversity for randomly sampled promoter sequence was 63% and that for coding region sequence was 25% . Highly conserved sequence would be plotted lower on the y-axis ( less diversity ) as displayed by the coding region line ( blue ) , while less conserved sequence would exhibit high diversity and appear near the top of the graph ( e . g . , gray , random promoter sequences ) . Small peaks observed in the promoter and coding region plots reflect degeneracy of the consensus sequence , with more degenerate positions exhibiting less calculated diversity . Patterns of the promoter and coding sequence lines are highly similar across species in Figure 4 except for being shifted on the y-axis . This was an expected feature of the data since these control curves were plotted as the average result of 1 , 000 trials of sequence fragment sampling across the human genome . When examining the population statistic of a large number of fragments , the average coding or promoter region fragments will exhibit similar transversion and transition mutation rates across species , which are visualized in these patterns . Figure 4D and 4E demonstrate that the positional sequence diversity of NRF2 and NFκB sites mirrored coding region sequence diversity across all species , as expected from the previous conservation analysis ( Figure 3 ) . Figure 4C shows a similar effect , with p53 cell-cycle–related sites displaying low sequence diversity . Intriguingly , the apoptotic-related p53 binding sites ( Figure 4B ) showed levels of sequence diversity that often met or exceeded those of the background promoter sequences . In rabbit and dog these apoptotic p53 binding sites have diverged so much that they may have lost function or could be under positive selection for mutation . The differences in positional sequence diversity between the two p53 RE subgroups were all highly significant ( two-tailed paired t-test assuming unequal variances: dog = 1 . 4e−13 , mouse = 4 . 3e−12 , rabbit = 2 . 0e−17 , and rat = 2 . 0e−15 ) . These dramatic results indicate again that these different classes of p53 binding sites may be subject to widely dissimilar sequence constraints .
The wide variety of genes transcriptionally regulated by p53 highlights the pleiotropic role of this master regulatory protein in many different biological pathways . Here , we addressed the conservation of human p53 binding sites across several mammals commonly used as experimental models . Examining global alignments of established human p53 binding sites , we found that common comparative genomics methods do not generally enhance p53 binding site prediction , although they can for NFκB , NRF2 , and a subset of p53 target genes involved in cell-cycle regulation . This apparent lack of conservation for many functional human p53 binding sites suggests that regulation of the p53 response network may be fine tuned for the needs of each species . By comparing sequence conservation between two p53 pathways , we have detected differences in the evolution of their regulatory elements . In particular , numerous functional human p53 REs in apoptosis-regulating sites , as well as the surrounding local sequence , show little homology to rodent sequences , suggesting that this ontology may have been shaped by primate-specific selection pressures that have resulted in turnover ( loss or gain ) of binding sites . This is supported to some degree by the very high mean sequence identity for all p53 REs between human and chimpanzee or monkey ( Figure S1 ) . However , turnover cannot be easily addressed by the global alignment method . For example , a short species-specific rearrangement such as an insertion of a repetitive DNA element ( e . g . , SINE , LINE , etc . ) that contains a RE would not globally align and would show zero identity across species but might maintain functional response across species . There were also seven DNA repair-regulating p53 REs in our dataset . Although there were not enough p53 RE sites to perform a statistically significant analysis , the average percent identity to the consensus sequence was similar to that of the apoptosis-related subset for human-to-rodent comparisons . These findings are significant considering the efforts to functionally model human p53 responses in the mouse ( including cell cycle , apoptosis , and DNA repair ) [33–35] . Complex molecular events ( reviewed in [35] ) regulate both p53 levels and activity prior to the transregulation of cell-cycle arrest and apoptosis genes . This results in large increases in p53 availability for binding to REs . Presumably the strength of p53 binding to a given target sequence has the effect of tuning regulation of the components of the p53 network within a species . Our data suggest that regulation of some p53 pathways , including apoptotic and DNA repair genes , may differ between humans and other mammalian species . Not only are REs in apoptotic and other genes different from cell-cycle genes in rodents , but they appear to differ from rabbits and dogs as well ( Figure 4B ) . This unexpected excess of sequence diversity for apoptotic elements could be explained by recent positive selection in all of these species . Support for this comes from an evolutionary analysis of a functional , yet poor , consensus-matching p53 RE in the apoptotic gene PIG3 . This study revealed that PIG3 became p53 responsive only recently , during primate evolution [10] and is consequently only present and functional in apes and humans . A recent emergence of primate-specific apoptosis p53 RE sites could explain the large number of interspecific differences identified following alignment to their orthologous mouse , rat , rabbit , or dog sites ( Figure 4 ) . Dermitzakis and Clark [36] observed a similar phenomena while surveying a broad number of TFBSs and concluded that a large percentage of apparently functional human sites were not functional in rodents ( and vice versa ) . The authors suggest that loss and gain of TFBSs has been commonplace in both rodents and humans . On the other hand , the p53 protein itself has changed very little between species . Mouse and human p53 proteins are 85% identical and show equivalent transactivation of human apoptotic and cell-cycle REs in a yeast-based system [37] . The DNA binding domain of the p53 protein has near 100% homology across all mammalian species indicating strong purifying selection to maintain DNA binding function . Our data indicate that cell-cycle REs are also being maintained by purifying selection , while the evidence suggests that divergent positive selection has occurred among REs of apoptosis genes . The evolution of apoptosis-related p53 binding sites has strong biological plausibility , as it seems likely that such modifications could profoundly affect how a species responds to environmental stress and cellular damage . With exposure to DNA-damaging agents being a common environmental feature throughout mammalian history , selection pressure and the evolution of systems to maintain genome stability could be quite different in rodents and primates . For example , it was recently shown that the Spalax ( mole rat ) , which lives its entire life underground , has a p53 protein with a very limited ability to induce several well-known human apoptotic genes in reporter assays . It is , however , quite capable of transactivating cell-cycle arrest genes [38] . The adaptation of the Spalax p53 response to a dramatically different environment underscores how separate pathways jumpstarted by the same transcription factor can have distinct evolutionary signatures . Cross-species analysis of p53-regulated genes in relation to biological function is largely absent . Thus it is unknown whether any preservation of functionality in apoptosis-related p53 binding sites exists , or if divergence and positive selection have created uniquely primate response characteristics . We are currently evaluating how p53 RE variation across species affects binding and transactivation in a functional model system and have observed that some weak binding REs show high conservation ( D . A . Bell , unpublished data ) . Other aspects of p53 pathways may evolve , such as the proteins that regulate the availability or quantity of active p53 protein , so that the sequence and binding affinity of affiliated binding sites could be coevolving with such changes . This study makes several suggestions for computational analysis of p53 REs and regulatory sequence in general . Since these binding sites seem to be experiencing much short-term evolution and turnover , comparative genome analysis in a panel of old and new world monkeys , “phylogenetic shadowing , ” may be a promising direction to enhance prediction accuracy [1] . Secondly , among those p53 REs exhibiting high conservation , mutations or polymorphisms that alter such sequences may be significant [35] . A key practical point is that if comparative genomics methods are used to identify putative functional regulatory regions , one should ensure that the choice of comparative species data is relevant to the selection pressure on the motifs of interest . For the p53 pathway , predictions based on mouse- or rat-to-human will not only generate a large excess of false positives , but many bona fide REs will be missed . Overall , conservation analysis may be a convenient measuring stick for regulatory element function , but we have shown that it must be used with caution and may depend on the TFBS category being analyzed . A reduction in evolutionary conservation in p53 regulatory elements is likely due to species-specific selective pressures acting on the distinct biological differences among p53-regulatory pathways .
TFBSs with experimental support were located in the literature ( see Table S1 ) . Genome coordinates ( National Center for Biotechnology [NCBI] 35 . 1 , May 2004 release ) were located using BLAT [39] searches against the human genome within the UCSC genome browser ( Table S2 ) [31 , 40] . If a TFBS could not be found in the genome , it was removed from analysis , which left 83 , 21 , and 21 binding sites for p53 , NRF2 , and NFκB , respectively . The UCSC “multiz17way” conservation track [31 , 40 , 41] provided a 17-way multiple sequence alignment between current releases of the Homo sapiens genome and eleven additional mammals: Pan troglodytes ( chimpanzee , November 2003 ) , Canis familiaris ( dog , May 2005 ) , Mus musculus ( mouse , May 2004 ) , Rattus norvegicus ( rat , June 2003 ) , Macaca mulatta ( rhesus monkey , January 2006 ) , Monodelphis domestica ( opossum , June 2005 ) , Bos taurus ( cow , March 2005 ) , Echinops telfairi ( tenrec , July 2005 ) , Loxodonta africana ( elephant , May 2005 ) , Oryctolagus cuniculus ( rabbit , May 2005 ) , and Dasypus novemcinctus ( armadillo , May 2005 ) . This alignment set was used to find the corresponding locations of each TFBS within each genome . The accuracy of these alignment regions were manually inspected and verified by both confirming similar local gene organization as well as referencing independently generated paired human–mammal alignments ( UCSC tables netMm7 , netMonDom1 , netBosTau2 , netRn4 , netCanFam2 , netRheMac2 , and netPanTro1 ) [42 , 43] . Similarly , a random list of promoter and cDNA sequences were obtained from Ensembl ( http://www . ensembl . org ) by referencing a genome coordinate list of all known human protein-encoding genes ( version 35 . 1 ) with an Ensembl gene identifier [44] . For each gene , a coordinate range of length equal to the TFBS of interest was randomly picked in the ( a ) 3 , 500 bases 5′ to the gene start site and ( b ) within protein coding DNA sequence . Sites from these two lists of coordinates were randomly chosen to form a set of genome regions with the same number of members as each TFBS category: NRF2 ( 21 ) , NFκB ( 21 ) , p53 ( 83 ) , p53 apoptosis ( 29 ) , and p53 cell cycle/cell growth ( 23 ) . This process was repeated 1 , 000 times with replacement to capture the variance in the data . As with the known TFBSs , the mammalian multiple alignment data from multiz17way were retrieved for each of these promoter and cDNA sites . The placement of target genes into p53 subcategories was based on the grouping used in Qian et al . and an annotated literature search ( Table S1 ) . Pair-wise percent identities relative to the each TFBS consensus were calculated as the percentage of RE bases that were either ( a ) identical between the human and second genome or ( b ) mismatched but do not deviate from the consensus sequence ( Figure 1 ) . For p53 REs , the variable spacer region was not considered . The distribution of conservation for each TFBS set is shown in Figure 2 . The conservation of the randomly chosen sets of coding region and promoter sequences , which represent the high and low extremes respectively of human genome conservation for TFBS comparisons , was calculated in the same fashion , and average results per x-axis bin for 1 , 000 trials are shown in Figure 2 . p-Values describing the statistical difference between percent sequence identity means for each TFBS set ( Table 4 ) were calculated using an unpaired two-tailed t-test with the assumption of unequal variances . The ORs in Table 4 were calculated as OR = ad/bc , where a = mean conservation of the column element , b = mean conservation of the row element , c = 100 − a , and d = 100 − b . A list of CNC blocks , which represent the 5% most conserved portions of the human genome , were downloaded from the “most conserved” track of the UCSC genome browser . These were generated by constructing a phylogenetic two-state Hidden Markov Model [45] from the 17-way multiple alignment , which includes the human 35 . 1 genome release [31 , 41 , 44] . To generate the data in Table 5 , the coordinates of each TFBS set were intersected with the CNC blocks . Likewise , the average number of randomly chosen promoter regions ( 1 , 000 trials ) found within CNCs was determined . The probability of finding a greater fraction of CNCs in randomly chosen promoter regions over TFBS sites was calculated using a two-tailed binomial test . For the ROC curves in Figure 3 , the sensitivity ( true positive rate ) and 1-specificity ( false positive rate ) of TFBS prediction were calculated at each of 11 conservation thresholds . The true positive rate was the fraction of bona fide TFBSs with a consensus sequence percent identity above a given level . The false positive rate was calculated as the average number of sites ( 1 , 000 trials ) that fell above a conservation threshold ( consensus sequence percent identity ) in an equivalently sized set of random promoter sequences . For example , 83 TFBSs composed the p53 ROC curve . Therefore , sets of 83 random promoter sites , where each site was 20 bases in length , were used to estimate the false positive rate . ROC AUCs ( Table 6 ) and standard errors were calculated directly from the graphs using the trapezoid rule as described by Hanley and McNeil [46] . Positional sequence diversity for a TFBS , which is related to the inverse of sequence identity , was calculated as the percentage of human nucleotides ( nt ) at each position that had a nonconsensus mismatch or insertion/deletion event when compared to one of the four mammals ( Figure 4 ) . Each TFBS of a given class ( i . e . , TP53 ) and its alignment to a mammal ( i . e . , mouse ) was pulled from the multiz multiple alignments to produce a set of paired sequence alignments . For each RE member of this pool , we counted the number of times where the first position of the mammalian sequence in the alignment differed from the human nt due to either ( a ) a nonconsensus mismatch or ( b ) an insertion/deletion event . This count was divided by the total number of TFBSs in a group ( i . e . , 83 for all p53 REs or 21 for all NFκB REs ) to get the percentage of two-way alignments that differed at that position . This value is taken as the sequence diversity at that first position . This calculation was then performed for the remaining nts in the TFBS . Mismatches from human that did not alter the consensus motif did not increase the percent sequence diversity and were ignored . | The p53 tumor suppressor is a transcription factor that coordinates cellular responses to DNA damage and stress , initiating cell-cycle arrest or triggering apoptosis . Evolutionary conservation is often employed to separate the functional “wheat” from the nonfunctional “chaff” when identifying binding sites of transcription factors like p53 . We evaluated evolutionary conservation of 83 experimentally validated human p53 binding sites against mouse , rat , rabbit , and dog genomes , and similarly examined binding sites for two other transcription factors as controls , NRF2 ( nuclear factor [erythroid-derived 2]-like 2 nuclear factor ) and NFκB ( nuclear factor of kappa light chain gene enhancer in B cells ) , which direct oxidative stress and innate immunity responses , respectively . NRF2 and NFκB binding sites both exhibited high interspecific conservation , indicative of purifying natural selection , but surprisingly , human p53 response elements on average displayed a lack of conservation . Thus conservation is not useful in the prediction of p53 binding sites . After grouping p53 REs by gene ontology , we observed that binding sites in cell-cycle genes like CDKN1A displayed high conservation , while p53 binding sites in apoptosis and DNA repair genes showed an unexpected excess of diversity and very little homology with rodent sequences . Overall these results reveal divergent evolutionary pressure among the binding targets of p53 and suggest caution in generalizing about the similarity of regulation of the p53 pathway between humans and rodents . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"primates",
"rattus",
"(rat)",
"dog",
"computational",
"biology",
"evolutionary",
"biology",
"homo",
"(human)",
"genetics",
"and",
"genomics",
"mus",
"(mouse)"
] | 2007 | Divergent Evolution of Human p53 Binding Sites: Cell Cycle Versus Apoptosis |
In flat-faced dog breeds , air resistance caused by skull conformation is believed to be a major determinant of Brachycephalic Obstructive Airway Syndrome ( BOAS ) . The clinical presentation of BOAS is heterogeneous , suggesting determinants independent of skull conformation contribute to airway disease . Norwich Terriers , a mesocephalic breed , are predisposed to Upper Airway Syndrome ( UAS ) , a disease whose pathological features overlap with BOAS . Our health screening clinic examined and scored the airways of 401 Norwich terriers by laryngoscopy . Genome-wide association analyses of UAS-related pathologies revealed a genetic association on canine chromosome 13 ( rs9043975 , p = 7 . 79x10-16 ) . Whole genome resequencing was used to identify causal variant ( s ) within a 414 kb critical interval . This approach highlighted an error in the CanFam3 . 1 dog assembly , which when resolved , led to the discovery of a c . 2786G>A missense variant in exon 20 of the positional candidate gene , ADAM metallopeptidase with thrombospondin type 1 motif 3 ( ADAMTS3 ) . In addition to segregating with UAS amongst Norwich Terriers , the ADAMTS3 c . 2786G>A risk allele frequency was enriched among the BOAS-susceptible French and ( English ) Bulldogs . Previous studies indicate that ADAMTS3 loss of function results in lymphoedema . Our results suggest a new paradigm in the understanding of canine upper airway disease aetiology: airway oedema caused by disruption of ADAMTS3 predisposes dogs to respiratory obstruction . These findings will enhance breeding practices and could refine the prognostics of surgical interventions that are often used to treat airway obstruction .
Amongst dogs , brachycephaly describes the head conformation of many popular breeds including the Bulldog , French Bulldog and Pug . This trait is grossly characterised by the concurrent rostrocaudal shortening and mediolateral widening of the skull and is accompanied by skin folds of the face . The structural discordance between the reduced facial skeleton and its overlying soft tissues such as the wrinkled skin folds underpins these breeds’ iconic looks , but these artificially selected aesthetics are under increasing scrutiny for their association with health problems including breathing difficulties . It is thought that soft tissues of the upper respiratory tract such as the nostrils , nasal mucosa of the turbinates and soft palate do not scale proportionately with reductions in the midface skeleton [1] . Misconfiguration of respiratory soft tissue restricts airflow and increases negative pressure within the airway [1 , 2] . This predisposes brachycephalic dogs to Brachycephalic Obstructive Airway Syndrome ( BOAS ) . Dogs diagnosed with BOAS can have stenotic nares , elongated soft palates and oversized , caudally protruding nasal turbinates [2–7] . Airway resistance caused by these tissue anomalies is believed to induce pathological remodelling of additional tissues including tonsil and laryngeal saccule eversion , oedema of the nasopharynx , laryngeal collapse , tracheal hypoplasia and exacerbation of the thickening and elongation of the soft palate [2 , 8 , 9] . Collectively , these perturbations severely impact the wellbeing of affected individuals by increasing their respiratory effort , resulting in laboured breathing , intolerance to heat/exercise , cyanosis and collapse [6 , 7] . The clinical assessment of the respiratory obstruction is often based on the grading of clinical symptoms , diagnostic imaging , and more recently , whole-body barometric plethysmography [6 , 10 , 11] . Treatment options for BOAS include anti-inflammatory medication which can reduce swelling/oedema acutely , however corrective surgery is often required to alleviate the condition [12] . Rhinoplasty of the nares , excision of the caudal aspect of the soft palate and aberrant turbinates , removal of the laryngeal saccules and tonsillectomy are the most common surgical procedures , which generally have mixed prognoses [2 , 3 , 6 , 12–14] . The number of patients requiring surgical treatment is expected to rise notably with the rapid increase in popularity of brachycephalic breeds . The costs and morbidity of surgical treatment are a welfare concern for both owners and their dogs , with complications reported in up to 25% and mortality in as many as 5% of cases treated surgically [14] . We and others have studied the underlying genetics of canine skull shape variation [15–19] . Variants in the BMP3 and SMOC2 genes are associated with canine brachycephaly , however the contribution of these variants to BOAS pathogenesis is unclear . Moreover , variants in both of these genes appear largely fixed among brachycephalic breeds that are at greatest risk of developing BOAS and yet the incidence and severity of BOAS differs between them [11 , 20] . BOAS heterogeneity may also be influenced by environmental and epigenetic factors , as well as other genetic modifiers segregating among dog populations . Under this premise , we became interested in the presentation of a respiratory condition remarkably similar to BOAS which has been identified in Norwich Terriers . As their name suggests , Norwich Terriers originate from south eastern UK where they were used for rodent control . Today , Norwich Terriers are recognised by all major kennel clubs and are known for their short , stocky build and prick ears . Dietschi et al . first described the presentation of Upper Airway Syndrome ( UAS ) in the Norwich Terrier . Although they are not considered a brachycephalic breed , affected Norwich Terriers present many of the hallmarks of BOAS including elongated and thickened soft palates , oedema of the nasopharynx and everted laryngeal saccules [21 , 22] . The closely related Norfolk Terrier , a breed that officially split from the Norwich Terrier in 1964 , is seemingly unaffected by UAS , suggesting genetic predisposition in the Norwich Terriers . Moreover , anecdotes from breeders regarding more recent dog generations , suggested that some Norwich Terriers appeared shorter-faced than those from earlier generations ( personal communication to JS ) . Indeed , Koch et al . postulated that selective breeding is driving Norwich Terriers to become brachycephalic [9 , 23 , 24] . Spurred on by these observations , and the possibility of uncovering genetic modifiers that increase respiratory obstruction risk , we sought to understand the genetic basis of Norwich Terrier UAS .
There is a continuum of head shapes observed across the domestic dog population ranging from the extreme brachycephalic to dolichocephalic conformations as represented by the profiles of the Pug and Smooth Collie , respectively ( Fig 1A and 1D ) . Respiratory tract disorders are markedly enriched amongst brachycephalic breeds such as the Pug , Bulldog , French Bulldog , Shih Tzu as well as the Norwich Terrier . The latter is not considered to be a brachycephalic breed , nor is the Norfolk Terrier ( Fig 1B and 1C ) . Rather both are generally considered “mesocephalic” . Indeed , linear measurements of the Norwich Terrier hard palate revealed intermediate palate dimensions between the extremes of facial morphology represented by the Pug and Smooth Collie ( Fig 1E ) . Furthermore , geometric morphometric analysis of the canine rostrum revealed that the Norwich Terrier occupies a morphospace distinct from classic brachycephalic breeds such as the Pug ( Fig 1F ) . For both linear measurements and geometric morphometrics , our data do not indicate a gross morphological difference between the Norwich Terrier and Norfolk Terrier . In 2000 , an upper airway screening programme for Norwich Terriers was established at the Vetsuisse Faculty of the University of Bern in Switzerland . The programme uses laryngoscopic videos to score ten components of the airway as normal , mild , moderate and severe ( S1 and S2 Movies and Table 1 ) [22] . An overall grade was given for the upper airway condition by combining the ten individual scores . Images taken from the laryngoscopic videos give examples of the soft palate length , laryngeal cartilage and laryngeal saccules graded as ‘normal’ ( Fig 2A–2C ) . In ‘severe’ graded examples , the soft palate is elongated and protruding caudally into the epiglottis ( Fig 2D ) , laryngeal cartilage is inverted into the lumen of the airway ( Fig 2E ) and the laryngeal saccules are everted ( Fig 2F ) . Histology from an unaffected Norwich Terrier reveals unremarkable connective tissue surrounding the laryngeal saccule ( Fig 2G ) whilst there is severe oedema and dilated lymphatic vessels in the connective tissue surrounding the laryngeal saccules of a ‘severely’ affected Norwich Terrier ( Fig 2H ) . To date , 401 Norwich Terriers in addition to 12 Norfolk Terriers were screened . Two-thirds ( 65 . 8% ) of Norwich Terriers had overall clinical presentations of UAS ranging from ‘mild’ to ‘severe’ whilst all Norfolk Terriers were unaffected ( Fig 3A ) . We selected 233 Norwich Terriers ( 109 male , 124 female ) representing phenotypic extremes of the phenotypic distribution for our genome-wide association study ( GWAS ) ( Fig 3B ) . Within this study population , everted laryngeal saccules ( 81 , 35% ) and elongated soft palates ( 33 , 14% ) were the most common severely-graded phenotype and were only graded as normal in 11 ( 5% ) and 43 ( 19% ) dogs respectively ( Table 1 ) . Meanwhile , cartilage stability ( 125 , 54% ) and trachea shape ( 112 , 48% ) were the most common normal-graded phenotypes . All unique phenotype pairings display a positive Pearson’s correlation coefficient ( range: 0 . 018 to 0 . 720 , median: 0 . 344 ) with the exception of the cartilage shape and oedema of the pharynx phenotypes ( r = -0 . 116 ) ( S1 Fig ) . Cartilage position and cartilage stability had the highest correlation of any phenotype pair ( r = 0 . 720 ) . Between 2003 and 2007 , The Swiss Terrier Club discouraged breeding dogs exceeding a ‘moderate’ upper airway phenotype . From 2007 onwards , upper airway screening was mandatory for all breeding pairs in Switzerland . As a testament to the coordinated efforts between veterinarians and breeders , the Swiss screening programme observed a reduction in the number of severely affected Norwich Terriers from 44 . 0% of those born between 1988–1997 to 8 . 6% for those born in 2006–2014 ( Fig 3C ) . The success of the screening programme underscores the heritability of UAS and suggests that the disease indeed segregates within this population . With cases of UAS reported across continents , the need to develop portable , cost-effective screening strategies became imperative . In order to further reduce the disease prevalence across the Norwich Terrier population and to provide insights into the pathophysiology of respiratory diseases that affect the upper airways of dogs , we sought to establish the genetic underpinnings of the condition . Genome-wide association analyses ( GWAS ) were performed for each of the ten upper airway phenotypes . Four phenotypes including eversion of the laryngeal saccule , oedema of the cricoid mucosa , oedema of the oropharynx and cartilage position returned markers with genome-wide significance ( Fig 4A , S2 Fig ) . The threshold for genome-wide significance was established by Bonferroni correction ( -log10 [0 . 05/105 , 130] = 6 . 32 ) . Regardless of phenotype , all association tests highlighted the same ~2 . 9 Mb quantitative trait locus ( QTL ) spanning 58 , 941 , 974–61 , 830 , 084 bp on canine chromosome ( CFA ) 13 with an index marker ( TIGRP2P185081_rs9043975 ) at 13:61 , 255 , 943 ( Fig 4B , S1 Table ) . Markers within this broad QTL display high levels of linkage disequilibrium ( LD ) ( r2 > 0 . 2 ) . Due to the modest correlation between individual traits ( S1 Fig ) , many markers ( 35/57 ) are significantly associated with at least two phenotypes ( S1 Table ) . Principal components analysis ( PCA ) of genotypes did not reveal phenotype-related substructure within the study cohort , adding confidence that the signal on CFA13 was truly associated with the disease ( S3 Fig ) . Genotypes extending ~1 Mb in both directions from the genome-wide significant markers on CFA13 were phased . Individual dogs were ranked by their disease severity and critical interval boundaries were defined by three meiotic recombinations . This revealed a 413 kb haplotype spanning chr13:61 , 166 , 179–61 , 579 , 985 that is shared among most severely affected Norwich Terriers ( Fig 4C ) . The disease-associated haplotype was homozygous in 75 . 3% ( 61/81 ) of severely affected dogs whilst it was homozygous in just 18 . 4% ( 28/152 ) of moderately-to-unaffected dogs ( S4 Fig ) . This critical interval spans the entirety of the ADAM metallopeptidase with thrombospondin type 1 motif 3 ( ADAMTS3 ) gene , in addition to ~114 kb and ~41 kb of sequence up and downstream of the gene , respectively . No other protein coding genes were annotated within the critical interval . To search for putative causal variants , we whole genome sequenced four Norwich Terriers representing the extremes of UAS phenotypes . This included two dogs that were homozygous for the CFA13 risk haplotype–one severely affected by UAS and the second seemingly unaffected . The remaining two dogs did not carry the CFA13 risk haplotype and were clinically unaffected . A total of 2 , 276 variants were called within the 413 , 806 bp critical interval and subsequently filtered ( see Methods ) , however no variants ( SNVs or indels ) were compelling candidates for causality based on location and/or interspecies conservation ( Table 2 , S2 Table ) . Following visual inspection of the whole genome sequences alongside aligned RNAseq data from a previous study [18] , we observed a gap in short-read coverage across all DNAseq and RNAseq datasets at exon 20 of ADAMTS3 suggesting an error in the CanFam3 . 1 assembly ( S5A Fig ) . We elected to generate a new local assembly for the CFA13 critical interval using long-read sequencing ( see Methods ) . DNA- and RNA-seq short reads were aligned to the new consensus sequence and revealed that exon 20 of ADAMTS3 extended an additional 133 bases beyond what was present in CanFam3 . 1 ( S5B Fig ) . Subsequent variant calling of the new 413 , 020 bp critical interval identified 1 , 834 variants . Variants were filtered based on allelic segregation between the disease-associated and alternate haplotypes , leaving a total of 80 single nucleotide variants ( SNVs ) and small indels . All remaining variants are in complete LD ( r2 = 1 ) . Two of the remaining variants are exonic–a synonymous variant in exon 21 and a missense variant in the newly defined exon 20 ( c . 2786G>A ) ( S2 Table ) . The missense variant is predicted to change an amino acid of ADAMTS3 from an arginine to histidine , p . ( Arg929His ) . This arginine is positioned within a thrombospondin type 1 repeat ( TSR1 ) domain and is invariable across mammalian species and close gene paralogs , suggesting evolutionary constraint ( Fig 5 ) . Accordingly , a substitution at this position is predicted to be “probably damaging” and “not tolerated” by PolyPhen-2 and SIFT , respectively [25 , 26] . We genotyped all Norwich Terriers and Norfolk Terriers screened in the study for the c . 2789G>A variant and did not observe the allele among the Norfolk Terrier population ( n = 12 ) , as expected , since UAS was not diagnosed in dogs of this breed . However , the risk allele was homozygous in 132 ( 32 . 9% ) and heterozygous in 195 ( 48 . 6% ) individuals from the Norwich Terrier cohort ( n = 401 ) . Dogs homozygous for the c . 2786G>A allele had a significantly greater total upper airway score than those heterozygous ( p = 9 . 10 x 10−17 ) or homozygous ( p = 3 . 08 x 10−20 ) for the ancestral allele ( Fig 6A ) . Seventeen Norwich Terriers were seemingly unaffected by UAS , two of which were homozygous for the c . 2786G>A risk allele and nine were heterozygous ( Fig 6A ) . Of note , many of the unaffected Norwich Terriers were young ( range 11 to 39 , median: 14 months old ) at the time they were screened . In contrast , the ADAMTS3 genotype does not segregate with weight , a suspected respiratory disease risk factor ( Fig 6B ) . Interestingly , by applying the Swiss Norwich Terrier club breeding guidelines to all 401 screened Norwich Terriers , 74 . 1% of those prevented from breeding are homozygous for the variant , whereas only 22 . 0% of Norwich Terriers permitted to breed have this genotype ( Fig 6C ) . Over 1 , 300 dogs representing up to 114 diverse breeds including representatives of brachycephalic breeds diagnosed with BOAS were screened for the c . 2789G>A variant ( S3 Table ) . The disease allele frequency ( AF ) was observed in the Norwich Terrier ( AF = 0 . 57 , n = 401 ) , Bulldog ( AF = 0 . 85 , n = 41 ) , French Bulldog ( AF = 0 . 12 , n = 23 ) , Staffordshire Bull Terrier ( AF = 0 . 125 , n = 8 ) , German Spitz ( Mittel ) ( AF = 0 . 06 , n = 8 ) and Pomeranian ( AF = 0 . 06 , n = 8 ) suggesting the variant may influence BOAS in the French and English Bulldogs . The 929His allele frequency in the human Exome Aggregation Consortium ( ExAC ) is less than 0 . 000017 [27] .
The incidence of UAS amongst the Norwich Terrier population presented a unique opportunity to identify disease modifiers that may be shared across brachycephalic and non-brachycephalic breeds alike . Leveraging laryngoscopic phenotyping , we identified and refined a QTL to an interval that encompasses a single positional candidate gene , ADAMTS3 . Following the correction of a local error in the canine reference sequence , we identified an ADAMTS3 c . 2786G>A missense variant that is associated with cases of UAS in the Norwich Terrier . Subsequently the French Bulldog and Bulldog , breeds susceptible to brachycephalic obstructive airway syndrome , were also identified as carriers of the ADAMTS3 missense allele . The ADAMTS proteins are a large family of protease enzymes [28 , 29] . The procollagen N-proteinases , which includes ADAMTS3 , are a subgroup of this family which were first shown to be expressed in cartilage amongst other tissues [30–32] . Within cartilage , ADAMTS3 has a substrate specificity for procollagen type II , which it cleaves to stimulate the maturation into collagen II , the major isoform of cartilage [33–37] . ADAMTS3 also has an important signalling function , as it proteolytically activates vascular endothelial growth factor-C ( VEGF-C ) , which in turn promotes lymphangiogenesis [38 , 39] . Loss of this signalling function in humans causes Hennekam lymphangiectasia-lymphedema syndrome 3 , a condition characterised by lymphedema and distinct facial features including hypertelorism and a flat nasal bridge [40–42] . This oedematous human phenotype is recapitulated in two different Adamts3 knockout mouse lines which were reported to have severe defects in lymphatic development [43–45] . Both knockout lines resulted in perinatal lethality with Ogino et al . , reporting death was due to apparent breathing problems . Interestingly , this line also presented with abnormal rib development and significantly rostrocaudally shortened skulls [43] . Whilst these studies did not specifically examine the tissues of the upper airways , the oedematous phenotype draws parallels with our observations in affected Norwich Terriers which carry the ADAMTS3 c . 2786G>A variant . Interestingly , in both the ADAMTS3 knock out mouse and cases of human Hennekam lymphangiectasia-lymphedema , craniofacial abnormalities are reported in conjunction with aberrant lymphatic development . Based on our analysis of rostra , we did not detect morphological differences between affected and unaffected Norwich Terriers , nor Norfolk Terriers . Similarly , there were no distinguishable differences in height or weight between the disease groups . Given the limitations of these assessments ( e . g . limited skull scans , imprecise postcranial measurements ) , it is possible that morphological differences between disease groups were undetected . Traditionally , the brachycephalic skull conformation has been considered the major predisposing factor to airway obstruction in brachycephalic breeds such as the ( English ) Bulldog and French Bulldog . The parallels in the upper airway oedematous phenotype in both brachycephalic dogs and the Norwich Terrier , along with the high prevalence of the ADAMTS3 c . 2786G>A variant in brachycephalic breeds raises the possibility that it promotes airway disease in these other breeds . This presents a new paradigm in our understanding of obstructive airway disease in that both a compromising skull conformation and a predisposition to oedema of the airway contributes to disease presentation . Complex genetic effects , which may include ADAMTS3 c . 2786G>A could explain the varying susceptibility to BOAS across brachycephalic breeds . The ADAMTS3 c . 2786G>A variant could not be separated from a further seventy-nine SNVs and small indels during filtering due to the long-range LD [46 , 47] . However , none of the additional intronic variants were compelling based on their position of cross-species conservation . Conversely , the arginine 929 residue in ADAMTS3 is highly conserved across orthologs and its close paralogs , ADAMTS2 and ADAMTS14 , from diverse vertebrates . Arginine 929 is located in the third of four thrombospondin type 1 repeats ( TSR1 ) within ADAMTS3 . The TSR1 repeats are thought to contribute to substrate binding and interactions with the extracellular matrix [48 , 49] . Although the variant site is outside the catalytically active metalloproteinase domain of ADAMTS3 , the canine Arg929His substitution might change or even disrupt the correct folding of the third TSR1 repeat . It has been shown that several arginine residues within the TSR1 domain contribute to the so-called central arginine layer , an important element of the three-dimensional structure of TSR1 repeats [50] . Thus , it is plausible that p . Arg929His might alter the functional properties of ADAMTS3 . The exact functional impact of the Arg929His substitution requires further investigation . The identification of ADAMTS3 in obstructive airway syndrome across dogs suggests a likely new role for the gene in effective respiratory function . This discovery warrants further longitudinal studies to assess possible correlations between the risk allele and complications during corrective upper airway surgery , where oedema of the upper airway can predispose dogs to post-operative complications . Identification of the ADAMTS3 c . 2786G>A risk allele is a critically important step to understanding the aetiology of airway disease , which at present is poorly understood . Future studies are warranted to understand the potential of the c . 2786G>A allele’s potential use as a diagnostic marker of disease .
All animal experiments were performed according to the local regulations . The dogs in this study were examined with the consent of their owners . The study was approved by the Federal Food Safety and Veterinary Office at the Federal Department of Home Affairs , Switzerland ( registration number 2 . 03 . 03 ) . Swiss biobanking was approved by the “Cantonal Committee For Animal Experiments” ( Canton of Bern; permits 22/07 , 23/10 , and 75/16 ) and the R ( D ) SVS Veterinary Ethical Review Committee ( 20 16 , University of Edinburgh ) . All animal experiments were performed according to the local regulations . The dogs in this study were examined with the consent of their owners . The study was approved by the “Cantonal Committee For Animal Experiments” ( Canton of Bern; permits 22/07 , 23/10 , and 75/16 ) and the R ( D ) SVS Veterinary Ethical Review Committee ( 20 16 , University of Edinburgh ) . A full description of the upper assessment was described previously [22] . In short , the upper respiratory tracts of 401 Norwich Terriers and 12 Norfolk Terriers were assessed in situ during endoscopic examination and scored retrospectively from video footage . All evaluations were conducted by a single veterinary surgeon . Subsequently , each of the ten phenotypic components of the airway ( soft palate length , soft plate thickness , laryngeal saccule , cartilage shape , cartilage stability , cartilage position , oedema of the oropharynx , oedema of the pharynx , oedema of the cricoid mucosa and shape of the trachea ) were scored on a range from 1 to 4 representing ‘normal’ , ‘mild’ , ‘moderate’ and ‘severe’ respectively by authors PS , ED and UR . A custom R script was used to generate Pearson’s correlation and dendrogram . Individual phenotype scores were weighted and summed to give the total airway score . Two-hundred and thirty-three Norwich Terriers ( 109 male , 124 female ) representing the extremes of upper airway phenotypes formed the study cohort . Participants varied in age from 7 to 188 months ( median = 18 months ) . H&E stains were done as previously described from dogs donated posthumously [51] . The geometric morphometric analysis of 3D skull reconstructions generated from computer tomography scans have been described previously [18] . Linear measurements of the hard palate were made , and the influence of allometry regressed using the neurocranium centroid size . PCA of the viscerocranium of 565 dogs representing 96 UK Kennel Club registered breeds permitted the comparison of face shapes . Whole blood samples were taken and stored in EDTA at 4°C prior to gDNA extraction following the whole blood protocol of the Nucleon BACC Genomic DNA Extraction Kit ( RPN-8502 , GE Healthcare Life Sciences ) . Genotypes were generated using the Illumina 170 , 000 SNV CanineHD bead chip by Edinburgh Genomics , UK and mapped to the CanFam3 . 1 coordinates . SNVs with minor allele frequencies < 0 . 05 and individuals with > 0 . 1 missing genotypes were removed using PLINK ( v1 . 90 ) [52] . Genotypes were imputed using a two-step process that included pre-phasing by SHAPEIT [53] and imputation by IMPUTE2 [54] . A total of 105 , 130 SNVs were used by GEMMA ( v0 . 94 . 1 ) in a univariate linear mixed model to perform GWAS [55] . A kinship matrix was implemented during the analysis with age and sex used as covariates . A Bonferroni correction threshold was used to determine statistically significant SNVs ( -log10 [0 . 05/105 , 130] = 6 . 32 ) . The LD of significant SNVs with all other markers in a 50 variant window was calculated using the independent pairwise test in PLINK ( v1 . 90 ) . Phased haplotypes encompassing the index SNV ( TIGRP2P185081_rs9043975 ) at chr13: 61 , 255 , 943 were ordered by UAS severity . The order was dictated by the four phenotypes returning significantly associated index SNVs ( laryngeal saccule > cartilage position > oedema of the cricoid mucosa > oedema of the oropharynx ) such that dogs with the most severe grade of all four phenotypes were positioned at the top . A consensus risk haplotype was the most frequent haplotype within the most severe scoring dogs , appearing in 98 of 162 chromosomes from severely affected dogs . Risk alleles were coloured based on whether they matched this consensus haplotype . The critical interval boundaries were defined by three or more meiotic recombination events across the severely affected Norwich Terriers . Four Norwich Terriers representing upper airway phenotypic extremes were whole genome sequenced to an average coverage of 15 . 9x . Two Norwich Terriers were homozygous for the disease-associated haplotype with one severely affected and the second apparently unaffected . The remaining two dogs did not have the disease-associated haplotype and were unaffected . DNA libraries were prepared using the TruSeq DNA PCR-free Library Preparation Kit . The Illumina HiSeq 2500 system sequenced 125 bp paired-end libraries with and average insert size of 419 bp . Reads from each resequenced Norwich Terrier were aligned to CanFam3 . 1 assembly using BWA-MEM [56] and variants within the critical interval chr13:61 , 166 , 179–61 , 579 , 985 were called for the CanFam3 . 1 and Zoey2 . 3 assembly using Platypus ( v0 . 8 . 1 ) [57] . Two Norwich Terriers homozygous for the disease-associated haplotype with differing phenotypes were selected with the potential of discovering the ancestral haplotype prior to the introduction of the causal variant ( s ) . To this end , for positions that had calls for all four Norwich Terriers , filtering criteria required variant ( s ) to be homozygous and exclusive to the affected dog , however this returned no variants . We hypothesised that the unaffected dog homozygous for the disease-associated haplotype was still subclinical due to the age of scoping at 1 . 2 years . Subsequently , filtering criteria required variants to be homozygous derived in the disease-haplotype carrying dogs and homozygous ancestral in those not carrying it . Norwich Terrier DNA and CanFam3 . 1-aligned RNAseq reads were viewed in IGV [58] which revealed an error in the reference sequence [18] . To resolve this error , we compared the local assembly of a Great Dane produced from PacBio long reads . In addition , we generated a de novo assembly from three bacterial artificial chromosomes ( BACs ) originally used for the CanFam3 . 1 assembly . BACs spanning the critical interval were sourced from the BACPAC Resource Center , Children’s Hospital Oakland Research Institute , California , USA ( CH82-24F19 , CH82-379O18 and CH82-101M10 ) . Following BAC isolation ( PhasePrep DNA Kit , Sigma-Aldrich , NA0100 ) , a DNA library was prepared from an equimolar mix of the three BACs for a single 1D barcode-free gDNA sequencing run using the Oxford Nanopore Technologies MinION platform ( SQK-LSK109 , R9 . 4 ) . A pipeline including Albacore ( v2 . 0 . 1 ) , Canu ( v1 . 5 ) and Nanopolish ( v0 . 8 . 4 ) was used to base call , construct contigs and improve consensus sequence respectively . The consensus sequences of both long-read platforms were in agreement and resolved the error underlying exon 20 of ADAMTS3 , though neither platform’s base calling resolved a ~40 bp intronic stretch of guanines downstream of exon 20 . Norwich Terrier short-read data ( European Nucleotide Archive study accession PRJEB16012 ) and RNAseq data ( European Nucleotide Archive study accession PRJEB17926 ) from a previous study were realigned to the new consensus using BWA-MEM and STAR ( V2 . 5 . 1 ) respectively with default parameters [18 , 59] . The RNAseq data was used solely to confirm exonic structure whilst the DNAseq data was used to repeat variant calling as previously described . Protein sequences of ADAMTS3 homologs ( HGNC:219 ) across species were downloaded from Ensembl and aligned using a ClustalW multiple alignment [60] . XP_539311 , a low-quality protein prediction of canine ADAMTS3 differed substantially from other species in its sequence corresponding to exon 20 –likely due to 133 bp of exon 20 missing from the CanFam3 . 1 assembly . To this end , the predicted amino acid sequence for the Norwich Terrier was created using comparative RNAseq alignment from nine dogs , representing eight breeds which were in full agreement for exon structure [18] . Residue positions are relative to the start codon . The thrombospondin-like domain ( PS50092 ) was predicted using PROSITE database of protein domains [61] . To genotype the ADAMTS3:c . 2786G>A variant , forward ( ACACACGAACCCAGGCACAC ) and reverse ( GGCCTGGGAGCACTGCAC ) primers were designed to amplify the region . PCR products were Sanger sequenced by Edinburgh Genomics , UK . All breeds used for genotyping were owner-reported . | Respiratory diseases are prevalent across dog breeds , particularly in brachycephalic breeds such as the Bulldog and French bulldog . The flat facial conformation of these breeds has long been assumed to be the major predisposing factor , however , the underlying genetics of their respiratory condition has never been elucidated . We became interested in the Norwich Terrier , a breed presenting with many of the same respiratory disease symptoms as the Bulldog . A distinction , however , is that the Norwich terrier is not considered to be a brachycephalic breed and so presented an opportunity to dissociate respiratory disease from head conformation . We performed a genome-wide association analysis for respiratory disease severity in the Norwich Terrier and resolved an association on chromosome 13 to a missense mutation in ADAMTS3 . Variants in this gene were previously shown to cause an oedematous phenotype–a disease characteristic in the airways of affected Norwich Terriers and brachycephalic dogs alike . We screened over 100 breeds for the ADAMTS3 variant and found that it is enriched in the Norwich Terrier , Bulldog and French Bulldog . This discovery changes how we view respiratory disease predisposition in the dog , offers potential genetic screens and highlights a new biological function for ADAMTS3 . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"animal",
"types",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"ears",
"vertebrates",
"pets",
"and",
"companion",
"animals",
"dogs",
"animals",
"mammals",
"palate",
"genetic",
"mapping... | 2019 | An ADAMTS3 missense variant is associated with Norwich Terrier upper airway syndrome |
Madariaga virus ( MADV ) , also known as South American eastern equine encephalitis virus , has been identified in animals and humans in South and Central America , but not previously in Hispaniola or the northern Caribbean . MADV was isolated from virus cultures of plasma from an 8-year-old child in a school cohort in the Gressier/Leogane region of Haiti , who was seen in April , 2015 , with acute febrile illness ( AFI ) . The virus was subsequently cultured from an additional seven AFI case patients from this same cohort in February , April , and May 2016 . Symptoms most closely resembled those seen with confirmed dengue virus infection . Sequence data were available for four isolates: all were within the same clade , with phylogenetic and molecular clock data suggesting recent introduction of the virus into Haiti from Panama sometime in the period from October 2012-January 2015 . Our data document the movement of MADV into Haiti , and raise questions about the potential for further spread in the Caribbean or North America .
Madariaga virus ( MADV ) , also known as South American eastern equine encephalitis virus , is an alphavirus in the family Togaviridae . Recent ecologic and genetic studies of eastern equine encephalitis virus ( EEEV ) have demonstrated clear separation between North and South American EEEV strains: North American EEEV cluster in a single genetic lineage ( lineage I , in the system proposed by Arrigo et al . [1] ) , with South American EEEV strains , or MADV , clustering in EEEV lineages II , III , and IV . MADV can cause outbreaks in horses , and appears to infect a variety of mammals , including rats and bats , and possibly birds and reptiles [1–3] . However , less than a dozen human cases of MADV infection have been documented , and almost all were encephalitis cases seen as part of an outbreak in Panama in 2010 [3 , 4] . In population-based serologic surveys in Panama and the Peruvian Amazon , between 2 and 5% of the general population had evidence of prior infection [2 , 3 , 5] , suggesting that mild or asymptomatic human infection is relatively common . In support of the latter hypothesis , we recently reported isolation of MADV from a child with acute febrile illness ( AFI ) , but no evidence of encephalitis , during the Zika virus ( ZIKV ) epidemic in Venezuela [6] . The virus has not been previously recognized in Hispaniola or other parts of the northern Caribbean . We report here the apparent recent introduction of MADV into Haiti .
The University of Florida IRB and the Haitian National IRB have approved all protocols , and written informed consent was obtained from parents or guardians of all study participants . Whole blood was collected into yellow top ( Acid Citrate Dextrose/ACD ) tubes ( Becton , Dickinson , and Company , Franklin Lakes , New Jersey ) , and an aliquot used to prepare blood smears for microscopic analyses for malaria parasites . To obtain plasma for virology analyses , a portion of the collected blood was centrifuged to pellet red and white blood cells , and the resulting plasma ( ca . 650μL ) transferred to sterile screw-top vials and stored at -80°C pending tests . As there was a possibility that viruses such as yellow fever virus or EEEV might be present in samples , RNA extractions and virology work were performed in the Lednicky BSL3 laboratory at the University of Florida’s Emerging Pathogens Institute , Gainesville , FL . RT-PCR tests for the detection of chikungunya virus ( CHIKV ) , dengue virus ( DENV ) , and ZIKV-genomic RNAs ( vRNAs ) in the plasma were accomplished as previously described [10 , 11] . Briefly , vRNA was extracted from virions in the plasma using a QIAamp Viral RNA Mini Kit ( Qiagen Inc . , Valencia , CA ) , and the extracted vRNAs tested using previously described RT-PCR primers [12–14] . To explore the possibility that the aforementioned viruses were present at levels too low to detect by RT-PCR in samples negative for CHIKV , DENV , and ZIKV vRNAs , or that other viruses were the causative agents , aliquots of plasma were inoculated onto monolayers of LLC-MK2 , MRC-5 , and Vero E6 as previously outlined [10 , 11] . In samples from eight patients with AFI , batches of inoculated cells formed virus-induced cytopathic effects ( CPE ) within 6 to 22 days that were reminiscent of the CPE observed for alphaviruses such as CHIKV: the infected cells developed dark , granulated cytoplasms with inclusion bodies , became enlarged , then either detached from the growing surface or appeared to undergo apoptosis ( Fig 2; virus strain list , and cell culture information , included as S1 Table ) . RT-PCR tests for the detection of CHIKV , DENV , and ZIKV were performed on vRNAs extracted from spent cell-media [10] , and all were negative for the viruses . Therefore , they were next tested with universal primer systems for the detection and identification of both alpha- and flaviviruses [15] . A very weak alphavirus amplicon was generated , though the putative alphavirus amplicon did not correspond in size to the alphaviruses identified by de Morais Bronzoni et al . [6 , 15] . Suspecting MADV , primers used in our previous work were used to screen samples by RT-PCR , and specific amplicons formed by primer pairs were sequenced [6] . At the same time , aliquots of spent cell media from four cell cultures that displayed advanced CPE were treated with cyanase nuclease ( RiboSolutions , Inc . , Cedar Creek , Texas ) , and vRNA thereafter extracted from the treated material [10] . Synthesis of complementary DNA was achieved as previously described [9] using non-ribosomal hexamers to favor the reverse transcription of viral genomes over ribosomal RNA [16] . PCR was subsequently performed with random hexamers and One Taq DNA polymerase ( New England Biolabs ) . Various prominent amplicons purified from a 2% agarose gel stained with ethidium bromide were sequenced . Both approaches revealed that the virus isolated was MADV . Sanger sequencing was performed on vRNA from spent cell lysates from four patients to obtain the MADV consensus sequences using methods similar to a previously published genome walking approach using overlapping primers ( S2 Table ) [6 , 9 , 10]; GenBank numbers are MH359230 , MH359231 , MH359232 , and MH359233 ( S1 Table ) . Because primers described in our previous MADV article [6] were suboptimal , the primer list in S2 Table depicts primers that were purpose-designed for work with these strains . All available MADV full genome sequences were downloaded from Genbank and codons aligned using MUSCLE [17] . Nucleotide substitution saturation and phylogenetic signal were assessed using DAMBE6 [18] and IQ-TREE [19] respectively ( see S1 Fig ) . Maximum likelihood ( ML ) phylogeny was inferred using IQ-TREE based on the best-fitting model ( GTR+F+G4 ) chosen according to Bayesian Information Criterion ( BIC ) . Strong statistical support along the branches was defined as bootstrap > 90% based on 2 , 000 replicates of Ultrafast Bootstrap Approximation [20] . The ML tree was used to check for temporal signal with TempEst [21] . A time-scaled phylogeny for the MADV isolates was then inferred with BEAST [22] v . 1 . 8 . 4 . by using the HKY85 nucleotide substitution model [23] , empirical base frequencies , and gamma distribution of site-specific rate heterogeneity . Strict and uncorrelated relaxed clocks , as well as constant size and Bayesian Skyline demographic priors were compared . The best-fitting model was chosen by calculating the Bayes Factor ( BF ) of marginal likelihood estimates ( MLE ) of different models , inferred with path sampling ( PS ) and stepping-stone sampling ( SS ) methods [22 , 24 , 25] . The strength of evidence against the null hypothesis ( H0 ) in favor of the more complex model ( HA ) , is evaluated according to the following guidelines: lnBF<2 no evidence; lnBF = 2–6—weak evidence; lnBF = 6–10—strong evidence , and lnBF>10 very strong evidence [25] . The best model for MADV isolates was strict molecular clock and Bayesian Skyline demographic prior ( S3 Table ) . A Markov Chain Monte Carlo ( MCMC ) sampler was run for 200 million generations , sampling every 200 , 000 , and proper mixing of the MCMC was confirmed when Effective Sampling Size ( ESS ) values for the parameter estimates were >200 using TRACER from the BEAST package . Maximum Clade Credibility ( MCC ) tree was extracted after 10% burn-in using Tree Annotator from the BEAST package .
The topology of the ML ( S2 Fig ) and MCC ( Fig 3 ) phylogenies were in agreement . As noted previously , EEEV isolates cluster within four lineages: Lineage I constitutes the North American EEEV strains , while MADV fall into lineages II , III , and IV ( Figs 3 and S2 ) . The new isolates from Haiti cluster within Lineage III , which comprises isolates from Central and South America , and separate the Central American isolates from the Southern American ones , forming a Central American-Caribbean monophyletic sub lineage . The time of the most recent common ancestor ( tMRCA ) for the sub lineage was 1939 , with a 95% High Posterior Density ( HPD ) interval of 1931–1948 . Within this new sub lineage the new MADV isolates from Haiti cluster close to Panama isolates collected in 2010 . The tMRCA for the MADV Haitian cluster was December 2013 , with a 95% HPD interval of October 2012- January 2015 , which corresponds to time window for the recent introduction of MADV in the island , likely from Panama . The evolutionary rate estimated for MADV was a 1 . 2 × 10−4 nucleotide substitution rate per year , in agreement with previous estimates calculated for this virus [1] .
Reports of human infection with MADV are rare , although cross-sectional serologic studies ( using plaque-reduction neutralization tests for confirmation ) in Panama and Peru have reported seropositivity rates in human populations of between 2 and 5% [2 , 3 , 5] , consistent with low-level MADV endemicity . Exposure may be substantially higher in epidemic settings , and/or with concurrent equine or animal epizootics: in recent work on seropositivity in household contacts of MADV and Venezuelan equine encephalitis virus cases during the MADV epidemic/epizootic in Panama in 2010 , 19 . 4% of household contacts were seropositive for MADV [26] . In that same study , it was also noted that seroprevalence was comparable in all age groups , as might be expected if the virus had been recently introduced into Panama [26] . The phylogenetic analysis of our Haitian strains is consistent with recent introduction of MADV into Haiti , while our report of isolation of MADV from a child in Venezuela documents ongoing transmission in that country , concurrent with a possible equine epizootic [6] . Taken together , these observations are consistent with ongoing transmission/emergence of MADV at multiple sites in the Caribbean and South and Central America . While most MADV case reports have involved patients with encephalitis , it is likely that the majority of infections occur in persons who are asymptomatic or who have only relatively mild disease [2 , 3 , 26] . This concept is supported by the previously noted work from our group in Venezuela [6] , with identification of MADV ( from a clade linked with Columbian and Venezuelan strains , distinct from our Haitian and Panamanian strains [Fig 3] ) in a 12 year-old girl with headache and fever , rash , and conjunctivitis , but no evidence of encephalitis . In our Haitian study we saw a similar pattern of symptoms and signs , but with only one child with a rash , and only one with conjunctivitis . Carrera et al , in their study in Panama [3] , used a pre-determined case definition for MADV of fever and headache . Only four of the eight patients in our study reported headache , suggesting that the Carrera case definition was overly restrictive . Interestingly , the pattern of symptoms in Haitian matched most closely with that previously reported from children in the same clinic with laboratory-confirmed DENV infections [8]; as with the DENV patients , the MADV patients were distinguished from the CHIKV patients by the lack of arthralgias . Overall , however , clinical presentation ( in the absence of meningeal symptoms and signs ) would appear to provide little assistance in diagnosing MADV infection . Culex ( Melanoconion ) pedroi has been identified as a primary enzootic vector for MADV in the Amazon Basin [27] . The virus was also recovered from Cx . ( Melanoconion ) taeniopus in an epidemic outbreak in Panama [28] , and in vector competence studies Aedes fulvus and Psorophora albigenu and Ps . ferox have been shown to be susceptible to and capable of transmitting the virus [29] . Cx . pedroi has not been previously identified in Haiti , but there are four known species in the Melanoconion subgenus including Cx . atrutus , Cx . carcinophilus , Cx . erraticus , and Cx . pilosus that are present [30] , together with Ps . ferox , which is known to be a very aggressive biter of humans . It remains to be determined if these native Melanoconion subgenus mosquito species and/or Ps . ferox serve as vectors for MADV in Haiti . In an extensive survey by Vittor and colleagues [2] of possible reservoir hosts in Panama , evidence of infection was only found in rat species , with the highest seroprevalence in the short-tailed cane rat ( Zygodontomys brevivauda; 8 . 7% seroprevalence , with one animal viremic for MADV ) and the black rat ( Rattus rattus; 3 . 9% seroprevalence ) . While Vittor found no evidence of infection in birds [2] , there are suggestions in earlier studies that birds and reptiles can also be infected [1] . In Haiti , no prior data are available on MADV in vectors , animal reservoirs , or humans . While we cannot exclude the possibility that MADV was present in Haiti before the current case cluster , our phylogenetic studies are strongly suggestive of recent introduction of the virus into Haiti from Panama . Recent work by our group [8–11 , 31 , 32] , and others , has underscored the apparent ease with which virus strains move among Caribbean and South and Central American countries . The drivers for this strain movement are varied . For Mayaro virus , we have shown a correlation between recent circulation of strains in this region and increased immigrant flow from Haiti to Peru and Brazil after the 2010 earthquake–and the counter-movement of peace-keeping troops from Brazil into Haiti during this same time period [32] . MADV is a little more complicated , as questions remain as to whether humans are a dead-end host , or if they can directly contribute to movement of the virus from one location to another . Over the past decade , there has been substantial movement of refugees from Haiti to and through Panama , as well as movement of Haitian workers to Panama; these population shifts may have provided an opportunity for movement of MADV from one country to the other . There is also the possibility that movement of strains was a function of movement of animal reservoirs ( such as the black rat ) on ships or in or on shipping containers; “hitch-hiking” of infected mosquitoes on airplanes is also a possibility [33] . At this point we know too little about the transmission and ecology of the virus to be able to predict its ability to move into other parts of the northern Caribbean , or areas such as Florida where North American EEEV is already endemic . Under these circumstances , the initiation of ongoing surveillance for MADV in humans , animals , and mosquitos throughout this region is clearly of public health importance . | Madariaga virus ( MADV ) is the name given to what used to be called South American eastern equine encephalitis virus ( EEEV ) , based on recent studies suggesting that MADV is distinct genetically from the EEEV circulating in North America . Until now , MADV has been found primarily in animals in South and Central America , with a limited number of human cases reported ( most of whom had encephalitis ) . Our group has been responsible for a series of studies assessing the etiology of acute febrile illness ( AFI ) among children in a school cohort in Haiti . Unexpectedly , in April , 2015 , we identified MADV on viral culture of plasma from a student with AFI in this cohort; an additional seven cases were identified on culture of samples from children with AFI in this same cohort in February , April , and May 2016 . On sequence analysis , all strains were very similar genetically , and appear to have come from a strain introduced into Haiti from Panama sometime in the period from October 2012- January 2015 . Symptoms of children were similar to those seen with dengue; none had encephalitis . Our data indicate that this virus , which has the potential for causing serious illness , has been recently introduced into Haiti , and raises the possibility that it might move into other parts of the Caribbean or North America . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"dengue",
"virus",
"children",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"education",
"togaviruses",
"eastern",
"equine",
"encephalitis",
"virus",
"pathogens",
"sociology",
"geographical",
"locations",
"micr... | 2019 | Emergence of Madariaga virus as a cause of acute febrile illness in children, Haiti, 2015-2016 |
Transmission of avian influenza viruses from bird to human is a rare event even though avian influenza viruses infect the ciliated epithelium of human airways in vitro and ex vivo . Using an in vitro model of human ciliated airway epithelium ( HAE ) , we demonstrate that while human and avian influenza viruses efficiently infect at temperatures of the human distal airways ( 37°C ) , avian , but not human , influenza viruses are restricted for infection at the cooler temperatures of the human proximal airways ( 32°C ) . These data support the hypothesis that avian influenza viruses , ordinarily adapted to the temperature of the avian enteric tract ( 40°C ) , rarely infect humans , in part due to differences in host airway regional temperatures . Previously , a critical residue at position 627 in the avian influenza virus polymerase subunit , PB2 , was identified as conferring temperature-dependency in mammalian cells . Here , we use reverse genetics to show that avianization of residue 627 attenuates a human virus , but does not account for the different infection between 32°C and 37°C . To determine the mechanism of temperature restriction of avian influenza viruses in HAE at 32°C , we generated recombinant human influenza viruses in either the A/Victoria/3/75 ( H3N2 ) or A/PR/8/34 ( H1N1 ) genetic background that contained avian or avian-like glycoproteins . Two of these viruses , A/Victoria/3/75 with L226Q and S228G mutations in hemagglutinin ( HA ) and neuraminidase ( NA ) from A/Chick/Italy/1347/99 and A/PR/8/34 containing the H7 and N1 from A/Chick/Italy/1347/99 , exhibited temperature restriction approaching that of wholly avian influenza viruses . These data suggest that influenza viruses bearing avian or avian-like surface glycoproteins have a reduced capacity to establish productive infection at the temperature of the human proximal airways . This temperature restriction may limit zoonotic transmission of avian influenza viruses and suggests that adaptation of avian influenza viruses to efficient infection at 32°C may represent a critical evolutionary step enabling human-to-human transmission .
Influenza viruses circulating in the human population are predominately type A and B , with type A being more common [1] . All influenza type A viruses originate from aquatic birds and successful introduction of these avian viruses into the human population , by either direct adaptation or reassortment with already circulating human viruses , has led to influenza pandemics of historical significance ( reviewed in [2]–[4] , [5] ) . Still , documented evidence of transmission of avian influenza viruses directly from birds to humans is rare , partly because species barriers restrict avian influenza virus infection of the epithelial cells of the human respiratory tract , the primary site of influenza virus infection and spread . Influenza A viruses possess a hemagglutinin ( HA ) attachment protein that binds sialic acid residues to facilitate infection of target epithelial cells . The HA of human influenza viruses preferentially binds to terminal sialic acid ( SA ) residues with α2 , 6 linkages , whereas avian influenza viruses preferentially bind to SA with α2 , 3 linkages [6]–[9] . The prevalence of α2 , 6 SA but paucity of α2 , 3 SA in the human respiratory tract has been considered to restrict infection by avian influenza viruses [10] . Recent reports , however , have detected significant levels of α2 , 3 SA on human airway epithelium both in vitro and ex vivo , including in nasopharyngeal and tracheobronchial tissue [11]–[14] . This SA distribution also correlated with avian influenza virus infection in vitro and ex vivo and raised the possibility that avian viruses could infect the upper airways in vivo . Therefore , although it is universally accepted that human-to-human transmission of avian influenza viruses requires adaptation of HA to switch from α2 , 3 to α2 , 6 SA usage , the cumulative data published to date indicate that SA linkages and their respective distribution in the human airways are not the sole barrier to avian influenza virus infection [15]–[17] . Other host factors and viral genes are likely also important determinants of infectivity . One such host factor that may limit zoonotic transmission is the difference in host temperatures between avian and human tissues that are susceptible to influenza virus infection . Avian influenza viruses are adapted for replication in the avian enteric tract at 40–41°C . While the surface temperatures of the human respiratory tract are variable , a temperature gradient clearly exists in which the surface temperature of the proximal large airways ( i . e . , nasal and tracheal ) average 32+/−0 . 05°C while temperatures of the smaller , distal airways ( i . e . , bronchioles ) are closer to that of the core body temperature , 37°C [18] , [19] . While multiple transmission routes have been described for influenza viruses , the proximal airways likely represent a predominant site for human influenza virus inoculation as they provide a large exposed surface area of virus-susceptible epithelial cells [20] . These cells are directly accessible by large droplet aerosols and by way of digital inoculation of the nasopharynx and conjunctival mucosa [12] , [21] . Inefficient infection by avian influenza viruses , even in the presence of α2 , 3-linked SA , may be due to the cooler temperature of the proximal airways compared to that of the distal airways/lung regions where H5N1 avian influenza viruses appear to replicate efficiently [22] . Avian influenza viruses are attenuated at temperatures below 37°C and cold sensitivity of avian viral RNA replication in cell lines was linked to the presence of a glutamic acid at amino acid 627 in the avian virus polymerase subunit , PB2 , instead of a lysine in the human virus PB2 [23] . Lysine substitution at residue 627 of H5N1 viruses improved virus replication in mice [24] . In addition to PB2 , work utilizing human-avian reassortant viruses in MDCK cells provided initial evidence that avian glycoproteins , HA and neuraminidase ( NA ) , may mediate temperature-dependent effects on viral growth [25] . To our knowledge , other viral genes have not been well characterized , nor the HA and NA further evaluated , in their contribution to temperature sensitivity of avian influenza viruses . To characterize the temperature dependency of avian vs . human influenza viruses in a relevant model of the target cell types of the human airways , we utilized an in vitro model of human ciliated airway epithelium ( HAE ) . This model closely mimics the morphological and physiological features of the human airway epithelium in vivo and has been previously used to investigate infection by diverse respiratory viruses [26]–[30] . In humans , ciliated airway epithelium is present throughout the airways , extending from the nasal cavity and large proximal airways into the distal bronchiolar airway regions . Previously , we have shown that both human and avian influenza viruses replicate well in HAE and that human and avian influenza virus cell tropism correlates with the respective distribution of the specific sialic acid linkages [13] . However , these previous studies were conducted at 37°C , reflecting conditions encountered in the distal airways [13] . Others have also utilized these airway cell systems to characterize influenza virus replication of wild-type and recombinant viruses at 35°C [14] , [31] , [32] . In the present study , we utilize the HAE model , in combination with influenza virus reverse genetics , to investigate the influence of temperature on human and avian influenza virus infection , replication and spread . We demonstrate that , compared to human influenza viruses , avian influenza viruses are severely restricted for infection of human airway epithelium at the temperature of the human proximal airways . Then , using different strategies to ‘avianize’ human influenza viruses , we show that the temperature restriction of avian viruses is closely associated with the avian HA and NA glycoproteins .
We and others have previously shown that human and avian influenza viruses infect and replicate in HAE [13] , [14] , [31] . Since our previous experiments were performed at 37°C , a temperature reflective of human distal airways , we have now compared human and avian influenza virus infection and growth in HAE at temperatures reflective of the proximal airways ( 32–33°C ) and distal airways ( 37°C ) . HAE were inoculated at either 32°C or 37°C with a low multiplicity of infection ( MOI; 0 . 01 ) of a representative human virus , A/Victoria/3/75 ( H3N2 ) , or an avian influenza isolate , A/Dk/Eng/62 ( H4N6 ) . Virus growth and spread throughout the epithelium at the two temperatures was measured and compared over time and infection further characterized with respect to virus-induced cytopathic effects ( CPE ) . At the temperature of the distal airways ( 37°C ) , the growth kinetics and mean peak titers of A/Victoria/3/75 and A/Dk/Eng/62 reached 2 . 3×108 pfu/ml and 4 . 7×107 pfu/ml , respectively , by 48 hours post-inoculation ( hrs pi ) ( Figure 1A ) . At the temperature of the proximal airways ( 32°C ) , A/Victoria/3/75 showed a modest delay in replication but still reached maximal titer of 7 . 8×107 pfu/ml by 48 hrs pi . In contrast , A/Dk/Eng/62 grew very slowly , with yields at time points up to 48 hrs pi reduced by 3 to 5 logs compared to growth for this virus at 37°C or A/Victoria/3/75 at either temperature . In comparison to 48 hr titers , A/Victoria/3/75 titers at both temperatures and A/Dk/Eng/62 titers at 37°C were reduced at 72 hr pi and every time point thereafter , indicating reduced progeny virus production . A loss of titer was also observed for A/Dk/Eng/62 at 32°C , but not before 120 hrs pi . To determine if loss of titer after reaching maximum levels correlated with increased CPE , we quantified adenylate kinase ( AK ) release by dead/dying cells into the apical compartment as a sensitive and global measure of cytotoxicity across the entire epithelial cell culture surface . Figure 1B indicates that substantial increases in AK levels , indicative of the onset of CPE , are first detected at 48 hrs pi for A/Victoria/3/75 at 32°C and 37°C and A/Dk/Eng/62 at 37°C . This induction of AK coincided with peak viral titer for these viruses under these conditions ( compare Figure 1A and 1B ) and suggested that the loss of titer correlated with the onset of CPE . Increasing levels of AK between 48 and 96 hrs pi were directly associated with continually decreasing viral titers , further supporting this claim . A relationship between the kinetics of virus growth in HAE and the level of CPE also suggested that CPE was a consequence of viral replication . This assertion is supported by the fact that trends in viral titers at a given time point are mirrored in AK levels detected 48 hrs later ( e . g . , compare viral titers at 48 hr pi ( Figure 1A ) to AK measurements taken at 96 hr pi ( Figure 1B ) ) . Since viral titer and AK levels could be related to the numbers of cells infected and/or the degree of virus replication within individual cells we compared titers of human and avian influenza viruses ( Figure 1A ) to the numbers of cells infected by each virus at the two temperatures over time . Immunodetection of viral antigen in inoculated HAE showed that human and avian influenza virus antigen was not detected 3 hrs pi , indicating that levels of antigen in residual viral inocula were below the limit of antibody detection ( data not shown ) . For A/Victoria/3/75 , a few isolated cells were positive for viral antigen by 6 hrs pi at 37°C , but by 24 hrs pi considerable numbers of antigen-positive cells were detected ( Figure 2A ) . In agreement with our growth curves in Figure 1A , A/Victoria/3/75 infected slightly fewer cells at 32°C compared to 37°C at 24 hrs pi , but importantly , A/Victoria/3/75 spread efficiently within the epithelium at both temperatures and differences in infection at early time points became less significant over time ( Figure 2A ) . In contrast to A/Victoria/3/75 , A/Dk/Eng/62 antigen was detected in only a few cells 24 hrs pi at either temperature . However , it should be noted that antigen-positive cells in en face images are viewed linearly ( Figure 2A ) whereas viral titers are shown on a logarithmic scale ( Figure 1A ) . Thus , an apparently small difference in titer as is seen at 24 hrs pi between A/Victoria/3/75 and A/Dk/Eng/62 at 37°C may correspond to a larger difference in the number of cells positive for viral antigen . While our staining also confirmed previous data that avian influenza viruses infect fewer human airway epithelial cells in comparison to human influenza virus at 37°C ( Figure 2A; [13] ) , the limited extent of A/Dk/Eng/62 antigen positive cells at 37°C by 24 hr pi was still unexpected given that titers at this time were slightly greater than those for A/Victoria/3/75 at 32°C . Whether this represents a difference in yield of infectious virus per infected cell between human and avian viruses is presently not clear . Overall , A/Dk/Eng/62 grew and spread well at 37°C , but was severely restricted for growth at 32°C and antigen positive cells were barely detectable before 48 hr pi for this virus at lower temperature . HAE cultures infected with A/Victoria/3/75 at either 32°C or 37°C and A/Dk/Eng/62 at 37°C viewed en face exhibited loss of integrity of the epithelium although the extent of injury and time of onset varied ( Figure 2A ) . Further evaluation of histological cross-sections indicated that A/Victoria/3/75 infection at 37°C , which had the highest and earliest induction of AK , resulted in the earliest evidence of morphological injury at 72 hrs pi . HAE infected with A/Victoria/3/75 at 32°C or 37°C or A/Dk/Eng/62 at 37°C all showed desquamation of the superficial layer of columnar epithelial cells with basal epithelial cells remaining attached to the matrix support by 120 hrs pi ( Figure 2B ) . Similar cytopathology has been reported for A/Udorn/307/72 influenza virus infection of HAE in vitro and for clinical human influenza virus infection in vivo [29] , [33] . The detection of AK in apical washes of A/Dk/Eng/62-infected HAE at 32°C suggested that this virus did eventually compromise cellular integrity at the lower temperature , but dramatic morphological effects were not seen at least for up to 120 hrs ( Figure 1B and 2B ) . It should be noted , however , that at 120 hrs pi , A/Dk/Eng/62-infected HAE at 32°C did display some morphological characteristics different from uninfected and infected HAE at earlier time-points . Preliminary assessment indicates that expansion of lateral spaces between the columnar epithelial cells had occurred . Although we do not know the significance of these morphological changes , we speculate these observations are the initiation of CPE that will ultimately result in similar cellular injury as seen for this virus at 37°C and human viruses at both temperatures . In sum , for both viruses at both temperatures , detection of maximal numbers of antigen-positive cells correlated with high titers ( compare Figure 1A and 2A ) and increasing CPE ( Figure 1B ) . By 72 and 120 hrs pi considerable loss of cells from the culture was evident and this correlated with the drop off in viral titers at these time points ( Figure 1A ) . Thus , we conclude that in the context of maximal infection in which there were no additional target cells available for infection within the finite surface area of the HAE culture , ongoing replication in antigen-positive cells shown at 48 and 72 hrs pi resulted in increased cell death . This CPE led to a reduction in the number of viable , virus-producing cells and in turn , to a reduction in progeny virus . Although A/Dk/Eng/62 induced CPE when sufficient titers were generated at 37°C , one consequence of restricted replication of this avian influenza virus at 32°C was a reduction in overt CPE in HAE , even at later time points associated with considerable viral titers . To determine whether other avian , but not human , influenza viruses display temperature dependent phenotypes , we performed multi-step growth curves with more human H3N2 isolates ( A/Eng/26/99 and A/Udorn/307/72 ) and A/Dk/Sing/97 , an avian isolate of different subtype ( H5N3 ) . Growth of both human-derived influenza viruses tested , A/Eng/26/99 ( H3N2 ) and A/Udorn/307/72 ( H3N2 ) , was not significantly different between 32/33°C and 37°C ( Figure 3A and 3B ) . Indeed , these two additional human influenza virus strains showed even less difference in titer between temperatures than was determined for A/Victoria/3/75 . Assessment of growth of avian influenza virus , A/Dk/Sing/97 ( H5N3 ) , over a 48 hr time course at 37°C showed similar growth kinetics to that of A/Eng/26/99 ( H3N2 ) , reaching titers of 7×105 pfu/ml and 1 . 6×106 pfu/ml , respectively ( Figure 3A and 3C ) . In contrast , at 32°C , A/Dk/Sing/97 ( H5N3 ) failed to grow at all ( Figure 3C ) . Clearly , the restriction of A/Dk/Sing/97 at 32°C compared to 37°C was an even more striking phenotype than A/Duck/Eng/62 . As the avian influenza virus strains used in this study were selected at random , with no selection for a temperature-dependent phenotype , we propose that low temperature restriction of avian influenza viruses , but not human influenza viruses , may be broadly characteristic of avian influenza viruses . The extent of restriction , however , may be variable between different virus strains . Since the avian virus isolates used in these experiments are neither derived from samples obtained from humans nor passaged in human cells in vitro , we next investigated whether growth attenuation at low temperatures would be retained in a highly pathogenic H5N1 ( A/VN/1203/04 ) influenza virus isolated from a fatal human case [34] . We compared infection kinetics of H5N1 ( A/VN/1203/04 ) at 33°C and 37°C on HAE using A/Udorn/307/72 in parallel cultures as a human influenza virus control . As described above , A/Udorn/307/72 grew with similar kinetics at 33°C and 37°C ( Figure 3B ) . A/VN/1203/04 , however , exhibited slower replication kinetics at 33°C when compared to that for 37°C ( Figure 3D ) . Indeed , titers were significantly decreased at 33°C vs . 37°C at 24 , 48 and 72 hrs pi . In addition , only at 37°C did A/VN/1203/04 approach similar peak titers as the human A/Udorn/307/72 virus by the end of the 72 hr time course ( Figure 3D ) . Histological analyses of A/VN/1203/04-infected HAE at either temperature showed absence of obvious CPE in sharp contrast to A/Udorn/307/72 that obliterated the epithelium by 72 hrs pi ( Figure 3E ) . The lack of obvious CPE after H5N1 infection contrasts reports that H5N1 induced extensive apoptosis in mammalian airway cells [35] , [36] . The fact that we did not observe obvious CPE with this highly pathogenic virus warrants further investigation but is in line with the limited cell damage shown following infection with A/Dk/Eng/62 for 72 hrs ( Figure 2B ) . In sum , using diverse examples of human and avian influenza viruses we have shown that avian influenza viruses , but not human influenza viruses , are restricted for infection and growth in HAE at the lower temperature of 32°C . Previously , the polymerase subunit PB2 has been shown to play an important role in host range restriction of avian influenza viruses in mammalian cells [37]–[39] . In influenza virus strains that circulate in humans , amino acid residue 627 in PB2 is a lysine , whereas in the majority of avian strains it is a conserved glutamic acid residue . The presence of glutamic acid at PB2 627 ( avian-like ) has been reported to account for the lower replication of avian influenza strains in mammalian cells and has been linked with reduced polymerase activity at lower temperature ( 33°C ) in some cell systems [23] , [24] . To assess the potential impact of this PB2 amino acid residue in restriction of avian influenza viruses at 32°C , we generated a recombinant A/Victoria/3/75 virus containing the PB2 K627E mutation and compared its growth with that of the isogenic wild-type virus in HAE at 32°C and 37°C . The K627E mutation resulted in restriction of the virus at both temperatures ( Figure 4Ai ) , and although titer at 32°C was 1 . 3 logs lower than at 37°C at 24 hrs pi , this difference was no greater than the differences in growth for wild-type virus at these temperatures ( 1 . 5 logs; Figure 4Ai ) . Moreover , at the later time points analyzed , 48 and 72 hrs pi , the PB2 mutant did not show a significant difference in titer between the two temperatures . These data indicate that the K627E mutant virus was restricted for growth in HAE but that restriction was not temperature-dependent . Indeed , quantification of the numbers of infected cells identified by en face staining revealed that the K627E mutant virus infected a similar percentage of cells compared to wild-type virus at 24 hrs pi ( Figure 4Aii ) and that the mutant was capable of spread since new cells were infected by 48 hrs with similar kinetics to that of wild-type A/Victoria/3/75 at both 32°C and 37°C ( Figure 4Aii ) . Statistically , there was no difference between the wild-type and PB2 mutant viruses at either 32°C or 37°C at 48 hrs pi with respect to percent influenza virus-antigen positive epithelium . Together , these data suggest that the amino acid residue at PB2 627 influences viral fitness in HAE , but does not confer to a human virus the temperature-dependent phenotype of avian influenza virus infection in human ciliated airway epithelium . Our initial phenotype indicated that A/Dk/Eng/62 was restricted in its ability to spread from cell to cell within the epithelium at 32°C ( Figure 2A ) . Several events in the viral life cycle that are critical for spread , including release of progeny virions from previously infected cells and attachment and entry into new target cells , are mediated by influenza virus glycoproteins . Thus , we hypothesized that glycoprotein function could be responsible for the restricted infection of HAE by avian influenza viruses at the lower temperature of 32°C . To test whether HA and/or NA contributed to the restricted phenotype of avian influenza viruses at 32°C , we used reverse genetics to generate mutant viruses genetically altered to confer avian virus-like glycoprotein specificities on the A/Victoria/3/75 background . First , mutations in HA previously shown to switch sialic acid usage from α2 , 6 to α2 , 3 linkages ( L226Q , S228G ) [40] were introduced to generate the Vic-226-228HA virus . Second , we generated a reassortant virus in which the Victoria NA was replaced by that of the avian virus A/Chick/Italy/1347/99 to generate Vic+Chick N1 . We again compared virus replication and spread of the recombinant viruses to that of wild-type A/Victoria/3/75 at the two temperatures . As stated above , replication measured for the wild-type virus was slightly compromised at lower temperature , noticeable at 24 hrs pi . Restriction at this time point was also observed during infection of HAE with Vic-226-228HA , as it had been for the PB2 mutant virus . Specifically , a 2 . 5 log decrease in virus growth was determined for Vic-226-228HA at 32°C compared to 37°C at the 24 hr time point ( Figure 4Bi ) . However , unlike the PB2 mutant virus , the difference between replication at 32°C and 37°C for Vic 226-228HA was also significant at the 48 hour time point . Moreover , this mutant virus with avian virus-like sialic acid usage spread less efficiently than wild-type at 32°C so that by 48 hrs pi the number of virus antigen-positive cells was significantly different ( Figure 4Bii ) . In contrast , at 37°C , Vic-226-228HA infected similar numbers of cells as the wild-type virus by 48 hrs; indeed , the mutant virus was able to spread significantly more efficiently at the higher temperature ( Figure 4Bii ) . Similarly , the reassorted virus Vic+Chick N1 displayed a 2 log decrease in viral titer in HAE at 32°C compared to 37°C at 24 hrs pi . Although this difference was not appreciably greater than the difference in titer between temperatures for either wild-type virus or the PB2 mutant , Vic+Chick N1 , unlike wild-type A/Victoria/3/75 and Vic 627PB2 , maintained the ∼2-log difference in growth at 48 hrs pi ( Figure 4Ci ) , suggesting this virus was more restricted at the cooler temperature . Quantification of numbers of infected cells illustrated that , like Vic-226-228HA , Vic+Chick N1 was restricted for spread at 32°C which was significant at 48 hrs , but was capable of spread similar to wild-type A/Victoria/3/75 at 37°C ( Figure 4Cii ) . Together these data suggest that avianizing either the HA or NA glycoprotein of an otherwise human influenza virus limits spread and subsequent infection at 32°C compared to 37°C . We next generated a recombinant influenza virus containing both the 226-228HA and Chick N1 and tested infection and growth in HAE at 32°C and 37°C in comparison to wild-type A/Victoria/3/75 . At 24 hrs pi , the double glycoprotein-altered virus exhibited similar restriction as observed for the other viruses . Nonetheless , an overall evaluation of the double glycoprotein-altered virus suggested that as infection proceeded , this virus was profoundly restricted at 32°C compared to 37°C ( Figure 4Di ) , exhibiting >2 log reduction in titer at 48 hrs . Notably , titers for the wild-type virus differed by less than 0 . 5 logs between temperatures at this time point . Furthermore , the double glycoprotein-altered virus was still significantly restricted at 72 hrs pi when titers at 32°C were compared to those at 37°C . The level of restriction observed for the double mutant was greater than that observed for either virus containing each of these mutations/substitutions individually . Moreover , analysis of viral antigen positive cells at 72 hrs by en face staining of infected HAE indicated compromised spread of Victoria ( 226-228HA ) +Chick N1 which was more severe at 32°C than 37°C ( Figure 4Dii ) . Determination of CPE during these experiments revealed that the double glycoprotein-avianized virus only produced CPE at 72 hrs pi when experiments were performed at 37°C , whereas wild-type human virus produced CPE earlier and at both temperatures ( data not shown ) . These data are consistent with the levels of CPE observed for A/Dk/Eng/62 ( H4N6 ) and A/Victoria/3/75 ( H3N2 ) in our initial studies ( Figure 1B ) and suggest that altering the human virus glycoproteins to avian virus-like characteristics has profound effects on infection , spread and CPE in the environment of the human ciliated airway epithelium . One potential caveat of the recombinant viruses with avianized HA and/or NA utilized in our previous analysis was that they contained HA and NA pairs that had not co-evolved . To eliminate the possibility that the restriction we observed with these recombinant viruses was due to an imbalance between the activities of the surface glycoproteins that were not evolutionarily optimized , we next generated reassorted influenza viruses on a common genetic background , possessing human or avian glycoproteins with co-evolved pairings . This was achieved using human recombinant A/PR/8/34 ( H1N1 ) in which the wild-type H1 and N1 glycoproteins were replaced by the H3 and N2 glycoprotein pair from A/Victoria/3/75 ( generating PR8+Vic HA/NA ) or the H7 and N1 glycoprotein pair from A/Chick/Italy/1347/99 ( generating PR8+Chick HA/NA , previously termed RD3 ) [41] . Since we and others have shown differential cell-type tropism between human and avian influenza virus in HAE [13] , [14] , we next determined if avianizing the human virus HA by mutation or substitution ( in the presence or absence of an avian NA ) recapitulated the cell-type tropism exhibited by wholly avian influenza viruses in HAE . As shown by immunofluorescent detection in histological sections of infected HAE , PR8 containing A/Victoria/3/75 glycoproteins infected both ciliated and non-ciliated cells in HAE with a tropism similar to wild-type A/Victoria/3/75 ( Figure 5 ) . In contrast , A/Victoria/3/75 with two avian-like amino acid substitutions in HA and PR8+Chick HA/NA only infected ciliated cells , a tropism that was mirrored by wholly avian virus [13] , [14] . These data clearly show that the ciliated cell tropism of avian influenza viruses is dictated by properties of the viral glycoproteins . These results correlate with the known increased sialic acid binding preference of avian HA for α2 , 3-linked SA , and to the presence of α2 , 3-linked SA on ciliated cells in HAE [8] , [13] , [14] . Growth kinetics in HAE of PR8+Vic HA/NA and PR8+Chick HA/NA inoculated at equal MOI ( 0 . 01 ) revealed that PR8+Vic HA/NA infection and growth was efficient at both 32°C and 37°C ( Figure 6A ) . PR8+Chick HA/NA grew at 37°C to identical titers as PR8+Vic HA/NA at 32°C recapitulating our data obtained for wholly human ( A/Victoria/3/75 ) and wholly avian ( A/Dk/Eng/62 ) viruses . In contrast , PR8+Chick HA/NA was severely delayed in growth at 32°C and generated titers that were >2 logs less than titers obtained for this virus at 37°C at both 24 and 48 hrs pi . Indeed , PR8+Chick HA/NA , like A/Dk/Eng/62 avian influenza virus ( Figure 1A ) , was significantly restricted for growth at 32°C at 12 , 24 and 48 hrs pi compared to growth at 37°C and growth of PR8+Vic HA/NA at either temperature . As observed for wholly human and avian influenza viruses , peak titers were reached for PR8+Vic HA/NA at both temperatures and PR8+Chick HA/NA at 37°C by 48 hrs pi after which a decline in viral titer was apparent . Again , as noted in our observations with human and avian influenza viruses , the loss of viral titers with time correlated with the onset of CPE . While PR8+Chick HA/NA infection at 32°C did not result in substantial AK release until 96 hr pi , increased AK activity was detected in cultures inoculated with this virus at 37°C . AK activity measured in cultures at this temperature increased with similar kinetics and reached similar levels as AK measured in cultures inoculated with PR8+Vic HA/NA at either temperature . Furthermore , the kinetics of AK induction demonstrated that again , AK was consequential to viral replication and that , overall , CPE induced by reassortant viruses was reflective of CPE measured for human and avian influenza viruses . En face staining of HAE at 24 hr intervals after inoculation showed PR8+Chick HA/NA spread to additional target cells at 37°C at a rate similar to that of PR8+Vic HA/NA at 32°C and correlated with the titers measured for these two viruses under those conditions ( Figure 6C and 6D ) . At 32°C , however , PR8+Chick HA/NA spread was severely compromised and resembled the infection characteristics shown for A/Dk/Eng/62 ( H4N6 ) in Figure 2A . Thus , by replacing human glycoproteins with those from an avian virus isolate , we have recapitulated the effect of temperature on infection and growth kinetics as well as the degree of cytotoxicity produced by wholly avian influenza virus interactions in human ciliated airway epithelium . The relative contributions of reduced cell-cell spread and reduced CPE by avian-like influenza viruses at temperatures of the proximal airways to in vivo infection and pathology will , however , require further investigation .
We have performed comparative studies of the infection kinetics of human and avian influenza viruses in a model of human ciliated airway epithelium at temperatures reflective of the human proximal and distal airways . Our data show that avian and avianized influenza viruses are restricted for infection and growth in HAE at 32°C but not 37°C , while human viruses infect and grow efficiently at both temperatures . Based on these data , we suggest that while the warmer temperatures of the distal airways enable comparable infection by both human and avian influenza viruses , the cooler temperatures of the human proximal airways only support efficient and robust infection of the ciliated airway epithelium by human influenza viruses . We speculate that the observed restriction for avian and ‘avianized’ viruses in HAE would render avian influenza viruses more susceptible to innate and adaptive immune responses that limit pathogenicity in vivo . These results have significant impact on our understanding of why avian influenza viruses rarely undergo zoonotic transmission and why , when the rare human case does occur , that avian influenza virus infection and pathology manifest predominately in the warmer distal airways and lungs . The inability of avian influenza viruses to replicate efficiently at cooler temperatures has been linked to the viral polymerase subunit , PB2 [23] , [24] . In the present study , mutating position 627 in a human virus PB2 to an avian virus conserved residue resulted in growth restriction at both 32°C and 37°C , suggesting that this residue is important for general viral fitness in HAE , but is not responsible for the differences in infection seen at 32°C vs . 37°C . Two recent reports also found that viruses with 627E in PB2 were attenuated regardless of temperature in human bronchial epithelial cells and MDCK cells , respectively , although in other cell systems including human small airway epithelial cells , a temperature specific effect was found [24] , [42] . It should be emphasized that those studies were performed in non-differentiated epithelial cells unlike our studies that use human differentiated airway epithelial cells . We and others have previously shown that differentiated airway epithelial cell models enable discrimination of attenuated phenotypes of respiratory virus infection whereas non-differentiated cells do not [26] , [27] , [43] . In addition , we also show using HAE , that the H5N1 strain A/VN/1203/04 , which possesses a lysine at position 627 ( human adaptation ) , is still restricted for growth at 32°C , albeit less so than avian influenza viruses that have never infected humans . The attenuation in HAE of this H5N1 isolate which possesses a “human” amino acid at residue 627 in PB2 suggests other residues in the polymerase subunit or other viral proteins altogether are involved in temperature sensitivity of avian influenza viruses . In our initial experiments , spread of avian influenza viruses from cell to cell at 32°C was compromised in cultures inoculated at low MOI , suggesting a potential role for the envelope glycoproteins , HA and NA , in mediating temperature restriction . Previous work by Kaverin and colleagues also demonstrated temperature effects on growth of human-avian reassortant viruses containing avian glycoproteins [25] , although this work was performed in non-polarized MDCK cells and did not investigate additional correlates of infection such as spread and CPE . In our study , we generated recombinant influenza viruses based on the A/Victoria/3/75 or A/PR/8/34 genetic backbone that were engineered to contain avian-like and/or avian glycoproteins and characterized infection in HAE . Kinetic studies showed that although human influenza viruses that possessed avian or avian-like surface glycoproteins were modestly restricted compared to wild-type viruses at 37°C , these mutant viruses were able spread like wild-type viruses throughout HAE at this temperature . Wide-spread infection throughout HAE was even observed for viruses in which their endogenous HA was replaced or mutated to preferentially bind α2 , 3 SA , restricting tropism to ciliated cells . Efficient replication of Vic-226-228HA at 37°C in our studies corroborates previous work by Matrosovich and colleagues in which little effect of HA-specificity ‘switching’ on replication was noted unless a very low MOI ( 0 . 00004 ) was used for inoculation [44] . In contrast , Wan and Perez described more profound differences in replication in HAE at 37°C with recombinant viruses that differed only in their receptor specificity [31] . However , it should be noted that their recombinant viruses were based on an H9N2 avian strain that yielded relatively low titers , and their initial infections were performed at 35°C before incubating at 37°C [31] . Compared to 37°C , viruses with a preference for binding to α2 , 3 SA , including Vic-226-228HA , were restricted for growth and spread in HAE at 32°C . Notably , the H5N1 strain examined in this study also maintains preference for α2 , 3 SA binding [45]; thus , we may surmise that this characteristic of A/VN/1203/04 contributes to its attenuation observed in HAE . The contribution of α2 , 3 SA usage to replication of influenza viruses investigated by Hatta et al . in the upper respiratory tract of mice may have been masked in the mouse model ( the 627 mutation in PB2 being more apparent ) as mice express solely avian virus-like receptors ( α2 , 3 SA ) in their airways [46] . Restriction of α2 , 3 SA-binding viruses in HAE at 32°C was not due to a discrepancy in SA expression since HAE maintained at either 32°C or 37°C expressed similar levels of α2 , 6 and α2 , 3 SA ( as detected by Sambucus nigra ( SNA ) and Maackia amurensis ( MAA ) lectin staining , respectively; data not shown ) . In conjunction with the HA , the sialidase activity of NA is crucial for successful virus penetration of mucus layers for initial infection and subsequent release of progeny virions from infected cells [47] , [48] . This is especially critical both in vivo and in HAE models in which the luminal epithelial cell surface is robust with glycoconjugates displaying abundant terminal sialic acid moieties that may act as false receptors for influenza viruses [49] . Using standard laboratory assays that employ small monovalent soluble substrates for cleavage by NA ( MUNANA ) , we were not able to demonstrate any temperature-dependent loss of NA activity associated with either human or avian virus ( data not shown ) . However , the ability of the avian virus NA to cleave biologically relevant substrates present in HAE may be compromised at 32°C vs . 37°C restricting both initial infection and subsequent spread of the virus throughout the epithelium . This is supported by our data which demonstrate restricted growth and spread of reassortant viruses containing avian virus NA , including Vic+Chick N1 and PR8+Chick HA/NA in HAE at 32°C . In addition to their independent functions , the balance between the binding affinity of the viral HA and the sialidase activity of the NA is also critical for efficient infection . The ability of A/Victoria/3/75 viruses with mutations or substitutions in either the HA or NA alone to infect similar numbers of cells and replicate to comparable peak titers as for wild-type virus at 37°C implies that these viruses were not crippled by the mismatch between the specificities of their HA and NA . Replication and spread of influenza viruses that possess an avian HA paired with its “matched” NA was even more compromised than that of recombinant viruses with individual changes to levels seen with wholly avian viruses . Thus , viruses with co-evolved glycoprotein pairs exhibit restricted replication at low temperatures and both HA and NA genes contribute to the phenotype . Together , these data imply that in the complex environment of the luminal surface of the human ciliated airway epithelium , the viral surface antigens have a marked effect on the extent of virus infection and that temperature plays an important role in limiting avian , but not human , influenza virus infection and spread in the cooler proximal airway regions . Given these results , we draw attention to other recently published data using the HAE model in which mutations in viruses that are growth attenuated in vivo display similar growth attenuation in HAE but not in non-differentiated cell lines , suggesting that HAE possess discriminating properties of attenuating phenotypes of mutants of respiratory viruses [26] , [27] . Admittedly , in the present study , despite restriction in both growth and spread , wild-type avian viruses and human viruses with avian or avian-like glycoproteins did eventually reach high titer at 32°C at later time points . The efficiency of infection and replication of a virus that inoculates the airway epithelium , however , is likely a critical factor in determining whether the virus is capable of establishing infection in a host that normally possesses innate and adaptive immune systems that attempt to limit virus infection and spread . At temperatures of the distal airways , avian influenza viruses displayed similar infection kinetics as human influenza viruses and would therefore , in the case of sufficient inoculum reaching these distal regions , be as likely to establish infection . Indeed , the clinical pathology findings for humans infected with H5N1 do report distal airway infection in ciliated bronchioles and lung regions [22] . Under these conditions of inoculation and infection , avian influenza viruses present in the distal airways may still be unable to spread to proximal airway regions without additional adaptation to cooler temperatures . One caveat of this prediction is that virus may be transported to proximal airway regions by innate mucus clearance mechanisms indicating that caution is required when attempting to identify proximal infection by viruses in airway secretions obtained from tracheal swabs . In conclusion , the present study substantiates differential host temperature as a critical barrier for infection by avian influenza viruses . Since the ciliated airway epithelium of the proximal airways is a major portal for influenza virus infection and spread , accessible by multiple inoculation routes ( e . g . , ocular , nasopharyngeal or aerosol ) , the inability of avian influenza viruses to establish infection and spread in these regions would be predicted to reduce the frequency of successful zoonotic transmission . Furthermore , the ability of human influenza viruses to generate high viral titers in the human proximal airways is likely a factor in effective human-to-human transmission and the induction of airway epithelial cell cytotoxicity as shown in this study may increase particulate matter perhaps associated with virus that facilitates inoculation of new hosts . Rapid induction of cytotopathic effects by human , but not avian , influenza virus infection at the temperature of the human proximal airways may also contribute to the onset of other host defenses such as sneezing and coughing that facilitate clearance of particulate matter/virus from the airways and potentially promote transmission between human hosts .
Human airway tracheobronchial epithelial cells isolated from airway specimens from patients without underlying lung disease were provided by the National Disease Research Interchange ( NDRI , Philadelphia , PA ) or as excess tissue following lung transplantation under University of North Carolina at Chapel Hill ( UNC ) Institutional Review Board-approved protocols by the UNC Cystic Fibrosis Center Tissue Culture Core . Primary cells derived from single patient sources were expanded on plastic to generate passage 1 cells and plated at a density of 3×105 cells per well on permeable Transwell-Col ( 12-mm diameter ) supports ( Corning , Inc . ) . HAE cultures were grown in custom media with provision of an air-liquid interface for 4 to 6 weeks to form differentiated , polarized cultures that resemble in vivo pseudostratified mucociliary epithelium , as previously described [50] . Madin-Darby Canine Kidney ( MDCK ) cells were maintained in DMEM ( Gibco-Invitrogen , Inc . ) supplemented with 10% fetal bovine serum and 1% penicillin / streptomycin ( Sigma-Aldrich , Inc . ) . Influenza virus A/England/26/99 ( H3N2 ) was isolated at the Health Protection Agency , Colindale , London , UK , during the routine surveillance program and has been minimally passaged in MDCK cells [51] . A/Dk/Singapore/97 ( H5N3 ) and A/Dk/England/62 ( H4N6 ) are typical avian influenza strains that have been passaged in both embryonated chicken eggs and MDCK cells during laboratory handling . Highly pathogenic A/VN/1203/04 ( H5N1 ) was biologically derived and minimally passaged in embryonated chicken eggs . A/Udorn/307/72 ( H3N2 ) was passed in baby hamster kidney ( BHK ) cells and represents a clone expanded once in embryonated chicken eggs . Recombinant viruses , including wild-type A/Victoria/3/75 ( H3N2 ) and mutants in either the A/Victoria/3/75 ( H3N2 ) or A/PR/8/24 ( H1N1 ) background , were generated from cloned cDNA in 293T and MDCK cell co-cultures as previously described [52] , [53] . Mutant viruses were generated in either the A/Victoria/3/75 ( H3N2 ) or A/PR/8/34 ( H1N1 ) genetic background as follows: 1 ) Vic 627PB2; A/Victoria/3/75 containing a lysine to glutamic acid amino acid substitution at position 627; 2 ) Vic-226-228HA; A/Victoria/3/75 containing two amino acid substitutions in the HA gene ( L226Q , S228G ) that confer an avian-like receptor binding preference [6] , [40]; 3 ) Vic+Chick N1; A/Victoria/3/75 in which segment 6 containing the endogenous N2 NA gene was exchanged for the N1 NA gene from avian isolate A/Chick/Italy/1347/99; 4 ) Vic-226-228HA+Chick N1; A/Victoria/3/75 containing both L226Q and S228G mutations and the avian N1; 5 ) PR8+Vic HA/NA; A/PR/8/34 in which the endogenous H1 and N1 were replaced with the H3 and N2 from A/Victoria/3/75 and 6 ) PR8+Chick HA/NA ( RD3 ) ; A/PR/8/34 in which the endogenous H1 and N1 were replaced with the H7 and N1 from A/Chick/Italy/1347/99 . ( RD3 was previously described as a candidate vaccine strain [41] . ) The last two reassortant viruses were generated by substituting segment 4 and segment 6 from PR8 with those from either A/Victoria/3/75 ( H3N2 ) or A/Chick/Italy/1347/99 ( H7N1 ) . The multi-basic cleavage site in the avian H7 HA gene used in these studies was removed prior to rescue of these recombinant viruses for safety . Available accession numbers ( GenBank: http://www . ncbi . nlm . nih . gov . libproxy . lib . unc . edu ) are V01086 for A/Victoria/3/75 HA and CAD37074 for A/Chick/Italy/1347/99 HA . HAE were rinsed with PBS to transiently remove apical secretions and supplied with fresh basolateral medium prior to inoculation . Virus inoculum was diluted in PBS and applied to the apical surface of HAE for 2 hrs at either 32°C , 33°C , or 37°C , as indicated . Following incubation , viral inocula were removed and cultures incubated at 32°C , 33°C or 37°C for the duration of the experiment . Viral growth kinetics were determined by performing apical washes with 300 µl of serum-free DMEM for 30 min at either 32°C or 37°C . Washes were harvested and stored at −80°C prior to analysis . Viral titers in the apical washes were determined by standard plaque assay or tissue culture infectious dose ( TCID ) 50 assay on MDCK cell monolayers as previously described [13] , [52] , [54] . At various points post-inoculation ( pi ) , HAE were fixed in cold methanol-acetone ( 50/50 ) and stored at 4°C . Cultures were then permeabilized with 2 . 5% triton-X 100/PBS++ ( containing 1 mM CaCl2 and 1 mM MgCl2 ) and blocked with 3% bovine serum albumin ( BSA ) in PBS++ before being probed with mouse anti-influenza virus nucleoprotein ( NP; Chemicon , Inc . ; 1∶100 ) and immunoreactivity detected with fluorescein isothiocyanate ( FITC ) -conjugated anti-mouse IgG secondary antibody ( Jackson ImmunoResearch Laboratories , Inc . , 1∶500 ) . Fluorescent images were obtained using a Leica DMIRB inverted fluorescence microscope equipped with cooled-color charge-coupled-device digital camera ( MicroPublisher; Q-Imaging , Burnaby , BC , Canada ) . The percentage of the epithelium positive for viral antigen as an index of percentage of infected cells was quantified over 5 images per culture by black and white pixilation of each image and computer calculation of percent black pixels after inverting the image . This technique determines percentage of black pixels in a defined area and does not account for differences in fluorescent intensity . Viral-induced cytotoxicity was determined by measuring adenylate kinase activity in apical washes using a commercially available assay ( Lonza , Inc . ) . Apical samples were centrifuged prior to freezing to remove any cellular contaminants present in the wash . Luminescence detected in samples from infected HAE were normalized to uninfected HAE and expressed as fold change over AK measured in uninfected ( mock ) HAE . Morphological assessment of cytotoxicity in HAE was performed with paraformaldehyde ( PFA , 4% ) -fixed histological sections ( 5 µm ) stained with hematoxylin and eosin . HAE maintained at either 32°C or 37°C for 72 hrs prior to sialic acid detection were washed , blocked with 3% BSA/PBS++ and probed with biotinylated SNA or MAA lectins to detect α2 , 6 and α2 , 3 SA , respectively ( Vector Laboratories , Inc . ; EY-Laboratories , Inc . ; 1∶100 ) . HAE were then fixed in 4% PFA and incubated with streptavidin-alexafluor 488 ( Molecular Probes , Inc . ; 1∶500 ) applied to the apical surface to detect lectin binding . HAE fixed in methanol∶acetone , were probed en face with antibody against viral NP ( Chemicon , Inc . ; 1∶100 ) and FITC-conjugated goat anti-mouse IgG1 and IgG2a ( Jackson ImmunoResearch Laboratories , Inc . , West Grove , PA; 1∶500 ) , then embedded in paraffin . Histological sections ( 5 µm ) were prepared and reprobed for viral antigen using standard immunofluorescence protocols . Briefly , sections were bathed in 2 . 5% triton-X 100/PBS++ for 30 min , blocked in 3% BSA/PBS++ and incubated with antibodies in 1% BSA/PBS++ . Primary antibodies were anti-viral NP ( Chemicon , Inc . , as above ) and anti-alpha acetylated tubulin ( Zymed Laboratories , Inc . ; 1∶2000 ) , a marker for ciliated cells . Secondary antibodies were FITC-goat anti-mouse IgG2a and Rhodamine red-conjugated goat-anti-mouse IgG2b ( Jackson ImmunoResearch Laboratories , Inc . ; 1∶500 ) . Sections were prepared with FluorSave mounting media ( EMD Chemicals , Inc . ) and images captured using a Leica DMIRB inverted fluorescence microscope equipped with a cooled color charge-coupled-device digital camera ( MicroPublisher; Q-Imaging , Burnaby , British Columbia , Canada ) . Linear mixed models were fitted to the repeated measurements of log-transformed viral titer over time that included effects for the four treatment groups ( defined by virus and temperature ) , eight time points , and the interaction between treatment and time . We note that in a small number of cases , there were only two treatment groups ( defined by temperature ) and fewer than eight time points . A heterogeneous autoregressive correlation structure of order one was assumed for the repeated measurements . A joint test of the interaction terms ( 21 degrees of freedom ) provides an assessment of the hypothesis of no differences among the four treatment groups with respect to viral titer growth ( log scale ) . Provided this test was significant , indicating some differences among the four growth curves , pair-wise differences between the three treatment groups versus the a priori specified reference group ( generally the avian strain at the lowest temperature ) were carried out for each time point , and significant differences at the 0 . 05 level were noted . No adjustments for inflated Type I error due to multiple comparisons were made . Missing observations were assumed to be missing completely at random , based on the fact that the investigators determined a priori to remove samples at specific time points during the experiment . | Influenza type A viruses are endemic in aquatic birds but can cross the species barrier to infect the human respiratory tract . While transmission from birds to humans is rare , the introduction of novel avian influenza viruses into immunologically naïve human populations has significant pandemic potential . Avian influenza viruses are adapted for growth at 40°C , the temperature of the avian enteric tract . However , the human proximal airways , the likely site of initial inoculation by influenza viruses , are maintained at a cooler temperature ( 32°C ) , suggesting that zoonotic transmission may be limited by temperature differences between the two hosts . Using an in vitro model of human ciliated airway epithelium , we show that avian influenza viruses grow well at 37°C , a temperature reflective of distal airways , but are restricted for infection at 32°C . A panel of genetically manipulated human influenza viruses possessing avian or avian-like surface glycoproteins were also restricted at 32°C , but not 37°C , suggesting that avian virus glycoproteins are not adapted for efficient infection at the temperature of the proximal airways . Thus , avian influenza virus infection is restricted in the human proximal airways due to the cooler temperature of this region , thus limiting the likelihood of zoonotic and subsequent human-to-human transmission of these viruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/viral",
"infections",
"virology",
"microbiology",
"infectious",
"diseases/respiratory",
"infections"
] | 2009 | Avian Influenza Virus Glycoproteins Restrict Virus Replication and Spread through Human Airway Epithelium at Temperatures of the Proximal Airways |
Pathogenic species of Leptospira are the causative agents of leptospirosis , a zoonotic disease that causes mortality and morbidity worldwide . The understanding of the virulence mechanisms of Leptospira spp is still at an early stage due to the limited number of genetic tools available for this microorganism . The development of random transposon mutagenesis in pathogenic strains a decade ago has contributed to the identification of several virulence factors . In this study , we used the transposon sequencing ( Tn-Seq ) technique , which combines transposon mutagenesis with massive parallel sequencing , to study the in vivo fitness of a pool of Leptospira interrogans mutants . We infected hamsters with a pool of 42 mutants ( input pool ) , which included control mutants with insertions in four genes previously analyzed by virulence testing ( loa22 , ligB , flaA1 , and lic20111 ) and 23 mutants with disrupted signal transduction genes . We quantified the mutants in different tissues ( blood , kidney and liver ) at 4 days post-challenge by high-throughput sequencing and compared the frequencies of mutants recovered from tissues to their frequencies in the input pool . Control mutants that were less fit in the Tn-Seq experiment were attenuated for virulence when tested separately in the hamster model of lethal leptospirosis . Control mutants with unaltered fitness were as virulent as the wild-type strain . We identified two mutants with the transposon inserted in the same putative adenylate/guanylate cyclase gene ( lic12327 ) that had reduced in vivo fitness in blood , kidney and liver . Both lic12327 mutants were attenuated for virulence when tested individually in hamsters . Growth of the control mutants and lic12327 mutants in culture medium were similar to that of the wild-type strain . These results demonstrate the feasibility of screening large pools of L . interrogans transposon mutants for those with altered fitness , and potentially attenuated virulence , by transposon sequencing .
Pathogenic Leptospira spp are the causative agents of leptospirosis , presumed to be the most widespread zoonotic disease [1] . Leptospirosis has emerged as a major public health burden in urban slums where risk is strongly linked to poverty and rat exposure [2–4] . It is estimated that there are 1 . 03 million cases and 58 , 900 deaths each year from leptospirosis [5] . The disease is transmitted to humans and animals through the urine of infected animals such as rats [1 , 6] . Bacteria enter the host via skin abrasions or mucous membranes and then disseminate via the bloodstream to target organs including the lungs , liver and kidneys . Infection produces a range of clinical manifestations , from flu-like symptoms to liver dysfunction , bleeding , kidney failure , pulmonary hemorrhage , and occasionally death [6 , 7] . The understanding of the virulence mechanisms of Leptospira spp is still at an early stage compared to other bacteria due to the limited number of genetic tools available for leptospires . The sequencing of a large number of leptospiral genomes [8] reveals that genes encoding proteins of unknown functions are enriched among pathogen-specific leptospiral genes [9] . Development of random transposon mutagenesis in pathogenic strains a decade ago has contributed to a better understanding of Leptospira biology [10] and has enabled identification of several virulence genes , including loa22 , the first leptospiral virulence gene to be described [11] , and lb139 , a gene encoding a potential signaling protein [12] . Interestingly , attenuation of virulence did not occur following inactivation of ligB or other genes whose products have been shown to have virulence attributes in vitro [13] . These results suggest a large degree of functional redundancy of virulence-associated gene products [6 , 14] . A disadvantage of virulence testing of individual transposon mutants is that this approach requires the use of a large number of animals . Although animal models remain critical for understanding leptospiral pathogenesis and for identifying virulence factors , there is a need for new approaches to reduce the number of animals required for such experiments . One strategy to minimize the number of animals is to inoculate pools of mutants into each animal . Recently , Marcsisin et al . screened pools of defined L . interrogans transposon insertion mutants for infectivity in the hamster model of acute infection [15] . 95 mutants were tested in pools of up to 10 mutants . 25 mutants were also tested in pools of 5 mutants in the mouse carrier model of infection . This study focused on whether mutants could be detected by PCR in cultures obtained from blood and kidney . Mutants with severe infectivity defects were tested individually for lethal virulence in hamsters . Only one mutant that failed to cause mortality was identified , although death was delayed with another four mutants . Because the level of each mutant was not quantified in the tissues , the approach was biased towards identification of highly attenuated mutants [15] . Transposon sequencing ( Tn-Seq ) has the potential to identify virulence-attenuated mutants with more subtle effects on infectivity . This technique combines transposon mutagenesis with the power of massive parallel sequencing . The basic principle of transposon sequencing methods involves DNA extraction from the pool of mutants , its cleavage by restriction enzyme digestion or sonication and the addition of adaptors for PCR amplification of the transposon ends and flanking regions . The PCR amplicons are analyzed by high throughput sequencing to determine the insertion site of each mutant and their relative abundance [16–18] . Tn-Seq has been used to study in vivo fitness of various bacteria such as Burkholderia pseudomallei [19] , Streptococcus pneumoniae [16] , Haemophilus influenza [20] , and Borrelia burgdorferi [21 , 22] and to identify bottlenecks during mouse infection by B . burgdorferi [23] . Tn-Seq has also been used to identify genes contributing to in vitro phenotypes , including antibiotic resistance in Pseudomonas aeruginosa [22] and carbon utilization in B . burgdorferi [21] . In the present study , we examined the potential of Tn-Seq to quantify the fitness of a pool of leptospiral mutants in various tissues during acute infection of the hamster . We screened mutants with transposon insertions in signal transduction genes to determine whether these genes affect the fitness of L . interrogans . The virulence of selected mutants with reduced fitness was tested in the hamster model of acute leptospirosis .
The pathogen Leptospira interrogans serovar Manilae strain L495 was used as the parent strain for generation of a transposon mutant library . The wild type ( WT ) strain and all mutants ( Table 1 ) derived from it were grown at 30°C in Ellinghausen-McCullough-Johnson-Harris ( EMJH ) medium [24 , 25] and EMJH supplemented with kanamycin ( Km , 50 mg/mL ) , respectively . Escherichia coli strain β2163 [26] containing the shuttle vector ( pCjTKS2 ) [27] , which carries a Himar1 transposon , was grown at 37°C in Luria broth supplemented with 2 , 3-diaminopimelic acid ( DAP , 0 . 3 mM ) , kanamycin ( Km , 50 mg/mL ) and spectinomycin ( Spc , 50 mg/mL ) . An L495 mutant library was generated by random transposon insertion mutagenesis [28] . Briefly , the shuttle vector pCjTKS2 , which contains a Himar1 element with its transposase gene lying outside of the transposon , was introduced into the L495 strain by conjugation with the E . coli β2163 donor strain . After two to three weeks of growth at 30°C on EMJH+Km plates , colonies were inoculated into EMJH+Km liquid medium and grown at 30°C for three to four weeks . Mutants were separately frozen at -80°C in EMJH and 4% glycerol ( final concentration ) without passaging . For each mutant , the insertion site of the transposon in the genome was determined by semi-random PCR as previously described by Slamti et al . [28] . PCR primer sequences are provided in the Table 2 . The insertion sites were identified by comparing the resulting sequence with the L . interrogans serovar Copenhageni Fiocruz L1-130 genome using the SpiroScope database ( http://www . genoscope . cns . fr/agc/mage ) [29] . All animals were routinely cared for according to the guidelines provided in the National Institutes of Health Guide to Laboratory Animal Care . Procedures involving hamsters were approved by the Veterans Affairs Greater Los Angeles Healthcare System Institutional Animal Care and Use Committee ( protocol #09018–14 ) . Hamsters were weighed daily and observed for endpoint criteria , including loss of appetite , gait or breathing difficulty , prostration , ruffled fur , or weight loss of >10% of maximum weight . Animals that met any of the endpoint criteria were euthanized by isoflurane inhalation followed by thoracotomy . Genomic libraries for sequencing were constructed as described by Troy et al . [23] . Genomic DNA was extracted from 100 μl of blood , 25 mg of tissue or frozen pellet from input pool with the DNeasy blood and tissue kit ( Qiagen , Valencia , CA ) following the manufacturer’s instructions except that an elution volume of only 100 μl was used . Extracted DNA was stored at -80°C until use . 50 μl of extracted DNA was sheared by sonication with a Fisher Scientific Model 505 Sonic Dismembrator for 3 min ( 10 s on and 5 s off; intensity , 80% ) in a high-intensity cup horn that was cooled at 4°C . Cytosine tails ( C-tails ) were added to 500 ng of sheared DNA using terminal deoxynucleotidyl transferase ( TdT ) ( Promega , Madison , WI ) . The TdT reaction mixture containing 475 μM dCTP and 25 μM ddCTP ( Affymetrix/USB Products , Santa Clara , CA ) was incubated for 1 h at 37°C followed by 20 min at 75°C . The DNA was then purified using the Qiagen MinElute PCR Purification kit ( Qiagen , Valencia , CA ) following the manufacturer’s instructions . The insertion site of the transposon was amplified by nested PCR . The first PCR was performed with 3 μl of the C-tailed DNA as template using olj376 and TnKN3 primers ( Table 2 ) specific for the C-tail and the Himar1 transposon , respectively , in a final volume of 25 μl . Primer olj376 , at the concentration of 1 . 8 μM , was added at three times in excess of TnKN3 ( 600 nM ) . Reactions were performed using DreamTaq Master Mix ( Thermo Scientific ) with an initial incubation of 2 min at 95°C followed by 24 cycles of 30 s at 95°C , 30 s at 60°C , and 2 min at 72°C followed by a 2-min extension at 72°C . The second PCR was performed with 2 μl from the previous PCR step with pMArgent2 primer ( 600 nM ) specific for the end of the transposon and an indexing primer ( 600 nM ) containing the specific sequences required for sequencing on an Illumina platform and a six-base-pair barcode sequence allowing all 37 samples to be multiplexed in a single sequencing lane ( Table 2 ) . PCR reactions were performed using DreamTaq Master Mix ( Thermo Scientific ) with an initial incubation of 2 min at 95°C followed by 18 cycles of 30 s at 95°C , 30 s at 60°C , and 2 min at 72°C followed by a 2-min extension at 72°C . PCR products were purified using a QIAquick PCR purification kit ( Qiagen , Valencia , CA ) following the manufacturer’s instructions except that an elution volume of only 30 μl was used . The majority of PCR products were between 200 bp and 600 bp in size . The DNA concentration of each culture and tissue library was measured with the Qubit 2 . 0 fluorometer ( Thermo Fisher ) . Equal amounts of DNA from each library were then pooled together and kept at -80°C until sequencing . The pooled libraries were sequenced on an Illumina HiSeq 2500 next generation sequencing system at the UCLA Neurosciences Genomics core facility as 64 bp single-end reads using the custom sequencing primer pMargent3 and the standard Illumina sequencing primer ( Table 2 ) . Data analysis was performed using the UCLA Galaxy platform [34–37] . Reads were cleaned by removal of ambiguous nucleotides , adapters , and primer sequences . The reads were filtered for length and quality: reads fewer than 20 nucleotides long or with a quality score of 20 or less for 95% of the cycles were eliminated . The remaining reads were mapped to the L . interrogans serovar Copenhageni strain Fiocruz L1-130 genome using Bowtie [38] . The resulting file was sorted to obtain a list of insertion sites , their corresponding gene numbers , and the number of reads per insertion site . In this way , the frequency with which each mutant occurred in each tissue and each animal was determined . Output/input ratios for each mutant were calculated by dividing a mutant’s output frequency by its frequency in the input pool . Output/input ratios across the 42 mutants were normalized by setting the median ratio for each animal to 1 . 0 . Ratios were compared to 1 . 0 ( neutral fitness ) using the Wilcoxon rank test with P values < 0 . 05 considered statistically significant . Comparison of ratios between duplicates was performed using the Student’s t-test with P values < 0 . 05 considered statistically significant . Correlations between the number of mapped reads and the load of bacteria in blood , kidney or liver were analyzed by the Pearson correlation test . Reproducibility of the Tn-Seq experiment was assessed using the Spearman correlation coefficient . Comparison of survival curves was performed using the Mantel-Cox log rank test . Comparison of motility ( diameter of growth ) between strains was performed using the Student’s t test with P values < 0 . 05 considered statistically significant . For all statistical tests , the number of asterisks indicates the significance level; * P < 0 . 05 , ** P < 0 . 01 and *** P < 0 . 001 . The number of bacteria in each sample ( Input pool , serum , kidney and liver ) was quantified with the Bio-Rad iQ5 real time system using the iTaq universal probe supermix . The lipL32 gene was amplified using the LipL32-45F and LipL32-286R primers and the LipL32-189P probe as previously described [30 , 39] ( Table 2 ) . The PCR mixture contained 250 nM of each primer , 150 nM of the specific probe , and 5 μl of DNA in a total volume of 20 μl . The amplification protocol consisted of 10 min at 95°C , followed by 40 cycles of amplification ( 95°C for 15 s and 60°C for 1 min ) . A negative result was assigned where no amplification occurred or if the threshold cycle ( CT ) was greater than 36 . Real-time PCR was performed in duplicate for each sample . Results were expressed as the number of leptospires/g of tissue used for DNA extraction or number of leptospires/mL of serum or culture . loa22::Tn , flaA1::Tn , ligB::Tn and lic20111::Tn and the WT strain were grown at 30°C in EMJH , supplemented with Km when necessary , to an OD420nm of ≈ 0 . 2 . Half of the ligB mutant and WT cultures were incubated with 120 mM NaCl for 4 hours at 30°C to maximize ligB expression [40] . Samples were separated on a 4–12% gradient NuPAGE Bis Tris precast gel ( Invitrogen ) and transferred to a PVDF membrane ( Millipore ) by semi-dry transfer at 25 V for 45 min with a Bio-Rad Trans-blot Semi-dry Transfer Cell unit . Membranes containing the loa22 mutant were probed with a 1/1 , 000 dilution of Loa22 rabbit polyclonal antiserum [41] and a 1/10 , 000 dilution of LipL41 rabbit polyclonal antiserum [42] . Membranes containing the flaA1 mutant were incubated with FlaA1 rabbit polyclonal antiserum at the dilution of 1/2 , 000 [43] and ImpL63 polyclonal antiserum at the dilution of 1/5 , 000 [44] . Membranes containing ligB mutant or lic20111 mutant were probed with mixture of a 1/2 , 000 dilution of LigAB rabbit polyclonal antiserum [45] and a 1/10 , 000 dilution of LipL41 rabbit polyclonal antiserum as a loading control . All membranes were then incubated in 1/5 , 000 dilution of horseradish peroxidase-conjugated donkey anti-rabbit immunoglobulin G ( Amersham ) and developed by enhanced chemiluminescence ( Pierce ECL reagent , Pierce ) . The motility of the flaA1 and lic20111 mutants in liquid EMJH+Km medium was analyzed by dark field microscopy and compared to the motility of the WT strain . Motility was also evaluated by spotting 0 . 5% agar semi-solid EMJH plates with 5 μl of four different cultures of the same mutant grown at the OD420nm of 0 . 2 . Plates were incubated for 15 days at 30°C , and diameters of growth were measured . Assays were performed in triplicate . The lic12327a , lic12327b , lic20111 , loa22 , ligB and flaA1 mutants and WT strain were cultured at 30°C in EMJH , supplemented with Km as appropriate . Growth was monitored daily by measurements of the optical density at 420 nm on a Pharmacia Ultrospec 2000 spectrophotometer . At least three independent growth curves were performed for each mutant and strain . The lic20111 mutant and WT strain were grown at 30°C to an OD420nm of ≈ 0 . 3 . Strains were cultured in duplicate . RNA was extracted from 20 mL of culture with Trizol reagent ( Invitrogen ) according to the manufacturer’s guidelines . Contaminating DNA was removed from RNA preparations using Turbo DNase from Ambion , and RNA was subsequently purified using the RNeasy kit ( Qiagen , Valencia , CA ) . 1 μg of each RNA sample was converted into cDNA with iScript Reverse Transcriptase Supermix ( Bio-Rad ) following the manufacturer’s instructions . The amounts of specific cDNA were determined by quantitative PCR using the Bio-Rad iQ5 real time system with the iQ SYBR Green Supermix as described [46] . Primer sequences are shown in Table 2 . The amount of cDNA of interest measured in each PCR assay was normalized to the amount of rpoB cDNA or flaB cDNA . The fold change of each gene was determined by the 2-ΔΔCt method [47] .
We created a library of over 800 Himar1 transposon mutants in L . interrogans serovar Manilae strain L495 . The insertion site in each mutant was determined individually by sequencing nested PCR products obtained by amplifying across one end of the transposon , as previously described by Slamti et al . [28] . An input pool of 42 mutants with insertions in 33 ORFs and 6 intergenic regions ( Table 1 ) was selected to validate the Tn-Seq approach and to examine changes in the composition of the mutant population across various tissues during infection . For validation of the Tn-Seq approach , the input pool included control mutants with transposon insertions in genes that have been tested for virulence in a rodent model of lethal leptospirosis . Insertional mutations previously shown to attenuate virulence occurred in loa22 , encoding a protein with an OmpA domain , and lic20111/lb139 , encoding a potential phosphatase that may modulate a two-partner switch mechanism controlling an alternative sigma subunit [11 , 12] . We also included mutants with insertions in genes previously shown to not be required for virulence including the adhesin gene ligB and the flagellar gene flaA1 [13 , 48] . Six intergenic mutants were included that harbored the transposon between open reading frames . To identify novel L . interrogans genes required for in vivo fitness , the pool included 23 mutants with insertions in 20 genes encoding putative signal transduction proteins . Our rationale for focusing on this gene category was that disruptions of signaling genes would be more likely to affect in vivo fitness due to their potential downstream effects on multiple functions . The input pool included two mutants with insertions in lic12324 encoding a gene containing a phosphatase domain , seven mutants with insertions in adenylate/guanylate cyclase genes , three in diguanylate cyclase or phosphodiesterase genes , five in histidine kinase genes , one in a gene encoding a member of the AcrR family of transcriptional regulators , and one in a gene encoding an alternative sigma factor ( Table 1 ) . The lic12324 , lic12327 and lic12627 genes were each represented by two mutants with insertions in different locations within the same open reading frame . We also selected ten additional mutants with potential roles in in vivo fitness , including mutants with insertions in lolD , encoding a homolog of the ATPase component of the lipoprotein export system , a heme oxygenase and the flagellar gene flaB2 ( Table 1 ) . Eight hamsters were challenged with the pool of 42 mutants . Four days post-challenge , blood , kidney and liver were collected . DNA extracted from these tissues and the input pool were analyzed by Illumina sequencing . 2 x 104–3 x 106 reads were obtained for each organ ( S1 Table ) . On average approximately 25% of the reads were discarded from the analysis during the cleaning phase of the sequence analysis . The remaining reads were mapped to the high-quality sequence of the Fiocruz L1-130 genome using Bowtie ( in the Galaxy software ) , and the frequency of each mutant within the bacterial population in each tissue and in each animal was determined . The nucleotide sequence of the Manilae L495 ORFs disrupted by the Himar1 element are 98 . 5–100% identical with those of the corresponding ORFs in the Fiocruz L1-130 strain . To examine the reproducibility of our Tn-Seq protocol ( sample preparation and sequencing ) , technical replicates with the input pool DNA were performed . Two sequencing libraries were created from the DNA with two different indexing primers . A strong correlation ( r2 = 0 . 9992 ) was observed between the composition of the population of mutants of these libraries ( Fig 1 ) , demonstrating the reproducibility of the amplification and sequencing methods . The same DNA preparations used for Tn-Seq were also used to quantify the total number of bacteria in each sample by TaqMan qPCR targeting the lipL32 gene . The results were expressed in terms of number of leptospires per gram of tissue or mL of serum ( S1 Table ) . The leptospiral load in the liver , ranging from 3 x 104/g to 2 x 108/g , was always higher than in the kidneys , where it ranged from 2 x 104/g to 1 x 107/g . In serum , the number of leptospires was lower: 1 x 104/mL to 3 x 105/mL . A significant positive correlation was found between the number of reads mapped and the burden of leptospires: r2 = 0 . 7963 ( P = 0 . 0029 ) , r2 = 0 . 7818 ( P = 0 . 0036 ) and r2 = 0 . 9068 ( P = 0 . 0003 ) in blood , kidney and liver , respectively ( Fig 2A–2C ) . Four days post-challenge , the composition of the population of mutants was quantified in blood , kidney and liver and was compared to that of the inoculum . The frequency of all mutants in the input pool was calculated , as well as their frequencies in each tissue and animal . The frequency of each mutant in the input pool ranged from 0 . 4% to 7% ( S2 Table ) . All mutants were detected in all tissues with changes in the composition of the populations . The means of the percentages in blood , kidney and liver ranged from 0 . 2% to 8% , 0 . 2% to 9% and 0 . 2% to 13% , respectively ( S2 Table ) . For each mutant , we calculated the output/input ratio , defined as the frequency of a mutant in the blood , kidney or liver divided by its frequency in the input pool . The output/input ratios across the 42 mutants were normalized to a median ratio to 1 . 0 in each animal . For each mutant and each tissue , we determined the median of the normalized ratios of the eight animals and compared it to 1 . 0 using the Wilcoxon signed-rank test . A fitness value of 1 . 0 is neutral , less than 1 . 0 is disadvantageous and greater than 1 . 0 is advantageous [49] . We observed statistically significant changes in fitness for 21 , 15 and 24 mutants in blood , kidney and liver , respectively ( Fig 3A–3C and S3 Table ) . In all tissues , eleven mutants had fitness values higher than 1 . 0 . In contrast , ten , four and thirteen mutants had decreased fitness in the blood , kidney and liver , respectively . A total of 12 mutants had statistically significant changes in all three tissues; with either decreased ( e . g . , the lic12327 and lic10203 mutants ) or increased ( e . g . , the lic12506 and lic13004 mutants ) fitness . In addition to the Tn-Seq results ( Fig 4 ) , we conducted Western blots ( Fig 5 ) , motility assays ( Fig 6 ) and growth curve analysis ( Fig 7 ) to confirm previously described phenotypes of the control mutants . To determine whether the virulence of the control mutants was as observed in previous studies , we assessed their virulence in a survival experiment ( Fig 8 and S4 Table ) and examined kidney colonization . Growth curves showed that none of the control mutants exhibited a defect in in vitro growth compared to the WT strain ( Fig 7 ) . In the pool of mutants used to challenge the animals , we included three pairs of mutants with different transposon insertion sites in the same gene: lic12324 , lic12327 and lic12627 . The insertion sites in the lic12324 gene were separated by 468 bp and by only 244 bp in the lic12327 gene . In the lic12627 gene insertions sites were farther apart ( separated by 910 bp ) ( Fig 9A ) . For the paired lic12324 mutants and paired lic12327 mutants , we observed a statistically significant decrease of in vivo fitness in both members of the pairs ( Fig 9B–9E ) . When we compared lic12324a::Tn to lic12324b::Tn and lic12327a::Tn to lic12327b::Tn , we did not see any statistically significant differences in their fitness . In the third pair of mutants , the fitness of the lic12627a mutant ( insertion site in the 5’ end of the gene ) was reduced in liver and kidney but significantly only in liver ( Fig 9F and 9G ) . However , the fitness of the lic12627b mutant ( insertion site in the 3’ end of the gene ) was significantly higher than 1 . 0 in blood and liver . The comparison of the ratios of these two mutants confirmed statistically significant differences in kidney ( P = 0 . 0379 ) and liver ( P = 0 . 011 ) . Previously , expression of the lic12327 gene has been shown to be upregulated by osmolarity [50] , suggesting its role in dissemination and survival in the host . Because of the decrease in fitness in both lic12327 mutants , we confirmed that neither mutant had a growth defect compared to the WT strain ( Fig 7 ) . We studied their virulence separately by challenging hamsters IP with 106 leptospires . Only one of six and two of six animals infected with lic12327b and lic12327a mutants , respectively , met endpoint criteria ( survival curves statistically different from the WT , P = 0 . 0019 and P = 0 . 0051 , respectively; Fig 8 and S4 Table ) . All animals had kidney colonization whether or not they met endpoint criteria . Twenty-three mutants with transposon insertions in putative signal transduction genes were included in this study ( Table 1 ) . Seven mutants did not present any change in any tissue . The sixteen other mutants exhibited in vivo fitness that differed significantly from 1 . 0: five mutants had changes in only one tissue , three in two tissues and eight in all three tissues ( Fig 3 and S3 Table ) . Among the eight mutants with changes in fitness in all tissues , five presented an increase in their fitness whereas three showed a decrease ( Fig 3 ) . The five mutants with an increase in their fitness in all tissues had transposon insertions in adenylate or guanylate cyclase genes ( lic12506::Tn , lic12670::Tn and lic13004::Tn ) , in a histidine kinase gene ( lic11432::Tn ) or in a transcriptional regulator gene ( lic13073::Tn ) ( Fig 3 and S3 Table ) . The three mutants with decreased fitness in all tissues had transposon insertions in an adenylate or guanylate cyclase gene ( lic12327a::Tn and lic12327b::Tn ) or in a phosphatase gene ( lic12324a::Tn ) ( Fig 3 and S3 Table ) . Thirteen mutants with transposon insertions in non-signaling genes were included in the pool of mutants ( Fig 3 and S3 Table ) . Three of these were “control” mutants and their behavior has been described above . Among the ten other mutants , lic11889::Tn , which has a transposon inserted in a flagellar protein , had increased fitness in all tissues and lic10203::Tn , which has an insertion in an epimerase gene , had diminished fitness . The fitness of the eight remaining mutants was not affected in any tissue . Six mutants with transposon insertions in intergenic regions were included in the input pool . Only one ( inter20138::Tn ) did not exhibit a change in fitness; two mutants ( inter10855::Tn and inter13722::Tn ) had increased in fitness in all tissues ( Fig 3 and S3 Table ) . The three other mutants showed decreased fitness in blood and liver . The distribution of the fitness values of the mutants differed among animals . We identified two types of distribution: a narrow distribution where all mutants have output/input ratios ranging from less than 2 . 5 log and a broad distribution where the range of ratios varies from more than 2 . 5 log ( S2 Fig ) . In blood and kidney , these two distributions are observed in the same animals: narrow distribution in animals 3 , 5 , 6 and 7 , broad distribution in animals 1 , 2 , 4 and 8 ( S2A and S2B Fig ) . In liver , only two animals have a narrow distribution , animal 3 and animal 7 ( S2C Fig ) .
We have developed a Tn-Seq assay to identify L . interrogans virulence genes candidates . The combination of transposon mutagenesis with the power of high-throughput sequencing successfully detected mutants with in vivo fitness defects . A major advantage of Tn-Seq is the ability to screen a large pool of mutants for altered in vivo fitness with a limited number of animals . This approach allowed us to reduce the cost of such an extensive screening of mutants , first by using a small number of animals and second by performing high throughput sequencing in a single lane in the Illumina system . Our findings with a small number of mutants suggest that Tn-Seq can be used as a first step to identify virulence genes of L . interrogans by screening large pools of mutants for defects in in vivo fitness . However , because not all L . interrogans mutants with diminished fitness within a pool of mutants will be attenuated in virulence [15] , experiments with larger numbers of mutants need to be done to better understand the relationship between fitness and virulence . This was the first Tn-Seq experiment performed with leptospiral mutants to assess their in vivo fitness . Due to the high bacterial tissue load , we were able to obtain sequencing reads from the DNA extracted directly from blood , kidney and liver . A culture step prior to DNA extraction was not necessary and allowed us to avoid in vitro growth bias . However , the approach can overestimate the relative abundance of a mutant by measuring the DNA from dead bacteria . We expect the contribution from dead leptospires to be minimal due to the exponential growth of the bacteria in tissues during the four days of infection [39] . We demonstrated the reproducibility of our protocol by processing the input pool DNA with two different indexing primers and comparing the frequency of each mutant obtained in both libraries ( Fig 1 ) . A strong correlation was observed between these libraries , demonstrating the reproducibility of our protocol ( PCR and sequencing ) . To validate Tn-Seq as a method to identify virulence genes of L . interrogans and to examine the relationship of fitness to virulence , the input pool included mutants with insertions in genes whose virulence had been examined in earlier studies . Transposon insertions in loa22 and lic20111 attenuated virulence [11 , 12] , whereas a transposon insertion in flaA1 and a targeted deletion of ligB had no effect [13 , 48] . In our experiments , no defect in fitness of either the ligB or flaA1 mutant was observed in any of the tissues tested ( Fig 4B and 4C ) and , as previously described , no attenuation in their virulence compared to the WT strain was seen in the hamster model ( Fig 8 ) . The loa22 mutant exhibited a decrease in in vivo fitness in blood and liver ( Fig 4A ) and attenuation in its virulence in the animal model ( Fig 8 ) . These results confirm previous studies showing a role for loa22 in virulence [11] . The partial virulence attenuation of the loa22 mutant differed from the findings of the study by Ristow et al . , in which the mutant was completely avirulent [11] . This difference may be related to the different parent strain: in our study , the loa22 mutant was generated in the highly virulent L495 strain ( LD50 < 102 ) [51] whereas it was previously obtained in the less virulent Lai strain 56601 ( LD50 > 107 ) [11] . Surprisingly , neither the in vivo fitness ( Fig 4D ) nor the virulence ( Fig 8 ) of the lic20111 mutant was affected , in contrast to the phenotype of another lic20111 mutant described in an earlier study [12] . The in vitro phenotype of our lic20111 mutant was similar to the WT in its motility ( Fig 6C and 6D ) , Lig protein production ( Fig 5D ) and growth ( Fig 7 ) , which is in opposition to the loss of virulence , reduction in motility , and diminished lig transcript production observed with an lic20111 mutant described by Eshghi et al . In that study , it was proposed that lic20111 is the first gene of a five-gene operon and that the insertion of the transposon in lic20111 caused attenuation of its virulence in the hamster model by polar effects on downstream transcription , which was verified by qRT-PCR [12] . We also observed diminished transcription of the downstream genes with our lic20111 mutant , although the effect was weak . The transposon is inserted in different locations in the lic20111 mutants: at the 3’ end of the gene in the Eshghi et al . study and near the middle of the gene in our study ( S1 Fig ) . The difference in virulence , motility , and lig expression between the two lic20111 mutants suggests that the location of the transposon within a gene may influence the mutant’s phenotype . By extension , the difference in in vivo fitness observed between the two lic12627 mutants ( Fig 9F ) may be explained by the different insertion sites of the transposon ( Fig 9A ) . This would suggest that the nearly full length LIC12627 protein generated from the lic12627b mutant , which harbors the transposon close to the 3’ end of lic12327 , retains adequate function to maintain the fitness of the strain . Although we lack experimental data that confirms expression of an active gene product from the lic12627b mutant , a similar effect was proposed by Lin et al . , [18 , 52] , who noted that in their collection of 4 , 479 transposon insertion mutants in B . burgdorferi , insertions in the last 10% of ORFs were over-represented . Five of the mutants included in our study have been studied previously in the experiment reported by Marcsisin et al . [15] , in which each animal was inoculated with a pool of 10 mutants . Similar results were obtained with some of our mutants . For instance , the lic12324 mutant , whose in vivo fitness was decreased in all tissues in our study , was detected by standard PCR in kidney and blood from only one and two animals out of five , respectively , in Marcsisin’s study . Comparable results were obtained in both studies with the lic13274 mutant , which was detected from four out of five animals in Marcsisin’s study and for which no change in fitness was observed in our Tn-Seq experiment . In contrast , while we observed no change or an increase in in vivo fitness of our lic20182 and lic10641 mutants ( Fig 3A–3C ) , these mutants could not be detected in blood and kidney ( except in the blood of one animal with lic10641::Tn ) in Marcsisin’s study [15] . These differences could result from differences in the inoculation dose , time of infection , or an unrecognized shortcoming with our assay . We identified two lic12327 mutants with reduced in vivo fitness in blood , kidney and liver ( Fig 9D and 9E ) . lic12327 encodes a putative adenylate/guanylate cyclase that contains a GAF domain and an adenylate/guanylate cyclase catalytic domain . This gene has been shown to be upregulated by physiological osmolarity [50] suggesting a role during host infection . The transposon insertion sites are 244 nucleotides apart from each other in these two mutants but both are located in the GAF domain . Both mutants were attenuated for virulence when tested individually in the hamster model ( Fig 8 and S4 Table ) and were recovered from kidneys . Experiments from another study demonstrated that a putative adenylate cyclase secreted from L . interrogans ( LA4008/LIC13201 ) elevated cAMP levels in a human monocytic cell line [53] . These observations suggest that cyclic nucleotides produced by L . interrogans play a variety of roles during infection . In contrast , we observed an increase in fitness for several mutants . The advantage of losing functional genes has been observed with Salmonella enterica , in which 25% of spontaneous deletions caused by randomly-inserted transposons caused enhanced growth rates under at least one of three growth conditions [54] . The enhanced fitness of mutants could be explained in part by the reduced metabolic burden in mutants that no longer synthesize proteins that are not essential for growth [54] . Alternatively , the functions provided by the disrupted genes may be provided by nearby mutants in the pool through the production of extracellular “common goods” [55] . For example , an inactivating mutation in the Pseudomonas aeruginosa gene encoding the quorum sensing regulator LasR results in the mutant out-competing the wild-type strain during co-culture [56] . Because signaling proteins are more likely that structural proteins to affect expression of multiple genes , our pool of mutants may be enriched for those with increase fitness in vivo . In addition to the control mutants that have previously been described , we included intergenic mutants in the input pool as potential controls for neutral effects on in vivo fitness . Surprisingly , only one out of six intergenic mutants was unaffected in fitness . Two of the intergenic mutants had an increase in fitness and three others exhibited a decrease ( Fig 3A–3C ) . These unexpected results can be due to the insertion of the transposon into a small RNA gene , into a promoter , regulatory or transcriptional terminator region , or into a protein-coding gene that has not been annotated . None of the small RNA genes identified in Camaino et al . ’s RNA-seq study were disrupted in our set of mutants [57] . Nevertheless , there are several general limitations to Tn-Seq that need to be considered . A decrease in fitness might be due to a mutant being out-competed by the other mutants in the pool rather than a direct effect of the transposon insertion on fitness [58 , 59] . An absence of change in in vivo fitness of a mutant that otherwise would have poor fitness in individual infections could be due to cooperation between mutants [60 , 61] . Indeed , 325 exoproteins have been identified in L . interrogans cultured under conditions mimicking infection [62] . Therefore , a mutant that fails to produce one of these exoproteins could be complemented intercellularly by another mutant . Additionally , bottlenecks during infection may impede recovery of random mutants from the original pool . However , all mutants from the input pool were recovered in all three tissues suggesting that bottlenecks did not significantly affect our experiment . Nevertheless , increasing the size of the pool may cause the stochastic loss of mutants during infection . This can be minimized by increasing the size of the inoculum [17] . Despite these limitations , we anticipate that Tn-Seq can be used to screen larger pools of L . interrogans mutants in a limited number of animals . | Leptospirosis is a neglected infectious disease that sickens many humans and animals throughout the world . It is caused by pathogenic Leptospira spp . Few leptospiral genes that contribute to the disease have been identified . We generated a library of 800 L . interrogans mutants with transposon insertions in different genes . Screening each mutant individually for the ability to cause disease in the hamster model would be laborious and requires thousands of animals . In a pilot experiment , we infected hamsters with a pool of 42 mutants to determine the role of the disrupted genes on fitness of the bacterium during infection . Out of the 12 mutants with diminished fitness in the three tissues , two had transposon insertions in the gene encoding an enzyme that may generate the small signaling molecule cAMP . When tested separately , the two mutants failed to sicken hamsters , indicating that intracellular cAMP signaling within L . interrogans could have a role in causing disease . These findings indicate that large pools of transposon insertion mutants can be screened in a limited number of animals to identify leptospiral genes that may be critical for the disease process . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Method",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"leptospira",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"population",
"genetics",
"microbiology",
"vertebrates",
"animals",
"mammals",
"gene",
"pool",
"genetic",
"elements",
"molecular",
"biolog... | 2016 | High-Throughput Parallel Sequencing to Measure Fitness of Leptospira interrogans Transposon Insertion Mutants during Acute Infection |
Previous studies have shown that stimulation of whole blood or peripheral blood mononuclear cells with bacterial virulence factors results in the sequestration of pro-coagulant microvesicles ( MVs ) . These particles explore their clotting activity via the extrinsic and intrinsic pathway of coagulation; however , their pathophysiological role in infectious diseases remains enigmatic . Here we describe that the interaction of pro-coagulant MVs with bacteria of the species Streptococcus pyogenes is part of the early immune response to the invading pathogen . As shown by negative staining electron microscopy and clotting assays , pro-coagulant MVs bind in the presence of plasma to the bacterial surface . Fibrinogen was identified as a linker that , through binding to the M1 protein of S . pyogenes , allows the opsonization of the bacteria by MVs . Surface plasmon resonance analysis revealed a strong interaction between pro-coagulant MVs and fibrinogen with a KD value in the nanomolar range . When performing a mass-spectrometry-based strategy to determine the protein quantity , a significant up-regulation of the fibrinogen-binding integrins CD18 and CD11b on pro-coagulant MVs was recorded . Finally we show that plasma clots induced by pro-coagulant MVs are able to prevent bacterial dissemination and possess antimicrobial activity . These findings were confirmed by in vivo experiments , as local treatment with pro-coagulant MVs dampens bacterial spreading to other organs and improved survival in an invasive streptococcal mouse model of infection . Taken together , our data implicate that pro-coagulant MVs play an important role in the early response of the innate immune system in infectious diseases .
Today it is generally accepted that coagulation is tightly interwoven with the innate immune system [1] . Both systems can act in a combined effort to sense and eradicate an infection in a highly sophisticated manner . Indeed , evolutionary studies suggest that fibrinogen has relatively recently acquired its function as a clotting factor because many fibrinogen-related proteins in invertebrates have an important role in defense processes , such as pathogen recognition , agglutination , and bacterial lysis , however , not in clotting [2] . This applies also to other members of the coagulation cascade , as sequence homology analyses in vertebrates revealed that many clotting factors share ancestry with complement proteases [3] . Together these results show that the vertebrate coagulation system has developed from evolutionary related cascades involved in innate immunity [4] . It is therefore tempting to speculate that coagulation has a yet underestimated function in the host defense to infection . The coagulation cascade can be broken down into an extrinsic ( tissue factor driven ) and intrinsic pathway ( contact activation ) . Both arms are initiated by limited proteolysis and are amplified in a snowball-like manner , eventually resulting in the generation of thrombin , which then initiates formation of a fibrin network [5] . The Gram-positive bacterium Streptococcus pyogenes is a major human pathogen that mainly causes local and self-limiting skin and throat infections . Infections can occasionally become invasive and develop into serious and life-threatening conditions such as streptococcal toxic shock syndrome ( STSS ) and necrotizing fasciitis . Notably , both conditions are associated with high morbidity and mortality ( for a review see [6] ) . The bacterium has evolved a variety of strategies to evoke activation of the coagulation cascade , involving for instance the induction of tissue factor on monocytes and endothelial cells by M proteins or an activation of the intrinsic pathway at the bacterial surface [7]–[9] . M proteins are streptococcal surface proteins and probably one of the best-known virulence determinants of this pathogen [10] . They can be released during infections [11] and act on monocytes to trigger cytokine induction and tissue factor up-regulation [8] , [12] . Recently we reported that soluble M protein triggers the release of pro-coagulant MVs from human peripheral blood mononuclear cells ( PBMCs ) . Once released from PBMCs these MVs can initiate coagulation by activating both pathways in a sequential mode of action [13] . Apart from PBMCs MVs can be secreted from almost all other human blood-born cells , and depending on their cell activation MVs can differ in their composition and function . Elevated levels of MVs have been related to pathological conditions such as bleeding and thrombotic disorders , cardiovascular diseases , cancer , and infectious diseases [14] . They form sphere-shaped structures , less than 1 µm of diameter and limited by a lipid bilayer . In contrast to their cell of origin , MVs from activated cells expose negatively charged phospholipids , mainly phosphatidylserine ( PS ) , on their outer membrane , which present a neo-exposed docking site for many plasma proteins including coagulation factors [15] . Despite an increasing knowledge on the role ( s ) of MVs in pathological processes e . g . as signaling molecules , in angiogenesis , and in initiation or propagation of coagulation and inflammation [14] , their function in infectious diseases is only poorly understood . In the present study we investigated whether pro-coagulant MVs are part of the innate immune response by exposing antimicrobial activity . To this end we performed a number of in vitro and in vivo experiments to show that pro-coagulant MVs not only efficiently prevent the proliferation of S . pyogenes bacteria within a formed clot , but also that application of human MVs in a subcutaneous murine infection model dampens bacterial spreading and improves survival .
PBMCs were isolated from human blood and stimulated with M1 protein as described in Methods . MVs were then purified as reported earlier [13] and the pro-coagulant activity of MVs was confirmed by measuring the clotting time ( data not shown ) . For subsequent binding studies , pro-coagulant MVs were tagged with gold-labeled annexin V and incubated with S . pyogenes bacteria in the presence of 1% plasma . Figure 1A depicts transmission electron micrographs at lower and higher magnification . At higher magnification the figure shows that pro-coagulant MVs are bound to the bacterial surface in the presence of plasma . To test whether the presence of MVs derived from other cells , interferes with the binding of pro-coagulant MVs from PBMCs , whole blood was stimulated with M1 protein . MVs were isolated and their binding to S . pyogenes was studied by transmission electron microscopy . Figure 1B ( upper panel ) shows that MVs isolated from M1 protein-activated blood bind to the bacterial surface . The origin of PBMC-derived MVs was confirmed by immunostaining with CD14 , also showing that activation of blood with M1 protein caused an increase in binding of monocyte-derived MVs ( Figure 1B , middle panel ) . To test whether the activation stage of the MVs contributes to binding , MVs were immunostained with an antibody against tissue factor . Figure 1B ( lower panel ) shows that only a few tissue factor-positive MVs were found attached to the bacteria , when MVs were isolated from non-stimulated blood . However , a more intensive antibody staining was recorded when MVs were recovered from M1 protein stimulated blood , showing that blood cell activation led to pro-coagulant MVs that bind to the bacterial surface . Based on these results we decided to use MVs isolated from PBMCs for all further experiments . MVs that were isolated from M1 protein stimulated PBMCs are therefore referred to as “pro-coagulant MVs” and from non-activated PBMCs as “ctrl . MVs” throughout the remaining part of this study . The interaction of MVs with S . pyogenes was further investigated by fluorescence microscopy . Pro-coagulant or ctrl . MVs were labeled with PKH26 ( red ) and incubated with S . pyogenes in human plasma . After a 30 minute incubation step , aggregates of MVs and bacteria ( DAPI-stained , blue ) were observed ( Figure S1 ) , similar to those described by Timár and colleagues [16] . The number of MV-bacterial aggregates that exceeded 10 µm was quantified ( Table 1 ) . The data show that both types of MVs bind and aggregate bacteria , but incubation with pro-coagulant MVs induced more and larger aggregates when compared with ctrl-MVs ( Table 1 ) . Next we tested whether opsonization of S . pyogenes with pro-coagulant MVs , renders the bacteria susceptible for clotting . To this end , S . pyogenes bacteria were pre-incubated with pro-coagulant MVs in the presence or absence of human plasma , washed thoroughly to remove non-bound MVs , and added to recalcified plasma . Under these experimental settings clotting occurred within 162 s as shown in figure 2A . If , however , bacteria were incubated with pro-coagulant MVs in the absence of human plasma , no clotting was observed within 300 s and likewise , incubation of bacteria with plasma in the absence of pro-coagulant MVs prevented clotting ( Figure 2A ) . Together the experiments imply that plasma protein ( s ) are required for the binding of pro-coagulant MVs to the bacteria and subsequent activation of clotting . Fibrinogen is a plausible candidate , as it is an abundant plasma protein and has high affinity for most streptococcal strains , including the AP1 strain , which was used in this study [9] . Therefore S . pyogenes bacteria were incubated with pro-coagulant MVs in the presence of normal or fibrinogen-depleted plasma , washed to remove non-bound MVs , and added to normal recalcified plasma . As before , when bacteria were pre-incubated with pro-coagulant MVs in the presence of normal plasma , clotting occurred within 169 s , while clotting was significant delayed ( 235 s ) when bacteria were pre-incubated with pro-coagulant MVs in fibrinogen-depleted plasma , prior re-calcification with normal plasma ( Figure 2B ) . Note that fibrinogen-depleted plasma was generated by defibrination and as fibrinogen was not completely removed ( 0 , 04 g/l are remaining ) , clotting was only delayed but not completely prevented . Previous work has demonstrated that M1 protein from S . pyogenes is the main fibrinogen receptor on the AP1 strain used in this study [17] . To test whether M1 protein is also the major fibrinogen binding protein that mediates the interaction between bacteria and MVs , we employed an isogenic AP1 mutant strain ( MC25 ) , which does not express M1 protein on its surface [18] . Wildtype AP1 and MC25 bacteria were pre-incubated with pro-coagulant MVs in the presence of human plasma , washed thoroughly to remove non-bound MVs , and added to recalcified plasma . As depicted in figure 2C , MC25 bacteria tagged with pro-coagulant MVs were not as potent to induce clot formation as AP1 bacteria . The number of MV-bacterial aggregates was quantified by fluorescence microscopy and also in these experiments we found that the MC25 strain was not as effective as the AP1 strain to form aggregates ( 5±2 vs . 49±11 ) in plasma when opsonized with pro-coagulant MVs . Finally we further investigated , whether other M proteins , either from the same serotype or from other serotypes , can recruit MVs to their surface . We therefore tested 14 clinical isolates , of which 5 were of the M1 type and 9 of other serotypes ( Figure S2A and B ) . When subjecting these strains to clotting assays we found that all serotypes had similar pro-coagulant activities as seen for the AP1 strain . Together the results show that the binding of pro-coagulant MVs to streptococci alters the bacterial surface from a non-coagulative to a pro-coagulative state . This interaction seems to be a common mechanism of group A streptococci , as also other serotypes explored similar clotting activities when incubated with pro-coagulant MVs . Moreover the data suggests that fibrinogen plays an important role in this chain of events . To study the role of fibrinogen as molecular bridge in more detail , surface plasmon resonance spectroscopy was employed . In a series of experiments we tested whether the activation state of MVs constitutes a regulatory mechanism that steers their affinity for fibrinogen . Sensor chips were coated with ctrl . or pro-coagulant MVs and probed with increasing concentration of fibrinogen . Though fibrinogen binding to both ctrl . MVs ( Figure 3A ) and pro-coagulant MVs ( Figure 3B ) was detected , determination of the association constants revealed that pro-coagulant MVs have a much higher affinity for fibrinogen than ctrl . MVs ( 0 . 019 nM vs . 3 . 3 µM , respectively ) as shown in figure 3C . The results from clotting experiments and fluorescent microscopy implicate an important role of M1 protein in binding pro-coagulant MVs ( see Figure 2C ) . To verify this conclusion we measured the interaction between M1 protein and pro-coagulant MVs , immobilized on a sensor chip , by surface plasmon resonance in the presence or absence of fibrinogen . Figures 3D+E illustrates that an interaction between M1 protein and the pro-coagulant MVs was only detectable when the chip was pre-incubated with fibrinogen , confirming fibrinogen's function as a bridging factor . In conclusion , the data show that MVs derived from activated cells expose additional binding sites for fibrinogen , which are required as docking sites for the streptococcal adhesion factor such as M1 protein or M proteins from other serotypes . In order to investigate how pro-coagulant MVs can up-regulate additional fibrinogen binding-sites , mass spectrometry analysis was used , which allows the identification and quantification of intracellular , membrane associated , and secreted proteins of MVs . With this approach a total number of 169 proteins , with a false discovery rate of 1% , was identified in non-stimulated and pro-coagulant MVs ( Table S1 ) . In ctrl . MVs , 57% of the proteins were cytosolic , 23% secreted , 12% membrane-associated , and 8% mitochondrial origin ( Figure S3 ) . This composition changed drastically in pro-coagulant MVs , as here an increase in secreted and membrane associated proteins was found ( 36% and 28% , respectively ) , while a decrease in cytoplasm and mitochondrial proteins to 35% and 1% was measured ( Figure S3 ) . We also noted a rise in the concentration of 34 proteins recovered from pro-coagulant MVs comparing to ctrl . MVs ( Table 2 ) . In particular , leucocyte elastase levels were dramatically up-regulated ( approximately 2500 times ) , but also higher levels of the fibrinogen-binding integrins CD18 ( 42 times ) and CD11b ( 7 . 8 times ) were noted . Another integrin , alpha-V/beta-3 , which is a receptor for a number of human proteins including fibronectin , laminin , and vitronectin were also found upregulated ( 2 . 9 times ) . Finally we noticed that proteins with antimicrobial functions such as lysozyme and neutrophil defensin 1 ( 3 . 7 times and 2 . 8 times , respectively ) were also enriched in pro-coagulant MVs . Taken together the determination of the protein content in ctrl . and pro-coagulant MVs by mass spectrometry analysis revealed that , apart from two fibrinogen-binding integrins , other proteins with an important role in the early immune response , are also up-regulated in pro-coagulant MVs . Recent studies support the concept that clot formation at the site of infection entraps bacteria in the fibrin network , which in turn prevents bacterial spreading , and promotes bacterial elimination [19] , [20] . Based on these reports , we speculated that MVs could also act as a clotting initiator that chains the bacteria within a formed clot . To prove this hypothesis , S . pyogenes were incubated in recalcified plasma followed by the addition of ctrl . MVs or pro-coagulant MVs . Artificial phospholipids with pro-coagulant activity ( PLs ) or tissue factor ( TF ) containing samples served as positive controls . Stable clots were formed when pro-coagulant MVs , PLs , or tissue factor were added to the bacteria/plasma mixture , while loose and less compact clots were generated when the bacteria/plasma mixture was incubated with buffer or ctrl . MVs ( not shown ) . The clot samples were covered with Tris-buffer containing 1% plasma and incubated for two or four hours at 37°C . Aliquots were collected from the supernatants and bacterial loads were determined . After two hours of incubation the number of released bacteria from plasma clots derived by pro-coagulant MVs was significantly decreased ( 9 . 7 times ) , when compared with the number found in the supernatants of samples incubated with ctrl . MVs ( Figure 4A ) . After the four-hour incubation , samples treated with buffer of ctrl . MVs contained high loads of streptococci . As seen before , incubation of bacteria in a plasma clot derived from pro-coagulant MVs prevented the escape of bacteria from the clots ( more than 12 times , comparing to ctrl-MVs ) and also PLs or tissue factor induced clots had a similar effect ( Figure 4B ) . These data demonstrate that bacteria are efficiently trapped and immobilized if they are opsonized with pro-coagulant MVs . It has recently been shown that activation of the coagulation cascade on the surface of S . pyogenes leads to an induction of antimicrobial activity [20] . To investigate whether antimicrobial activity is also seen when clotting is induced by pro-coagulant MVs , additional bacterial growth experiments were performed . Streptococci were mixed with plasma and clotting was initiated by adding pro-coagulant MVs , PLs , or tissue factor . Ctrl . MVs or buffer served as controls . After 30 min , clots were homogenized and bacterial loads determined . As seen in figure 5A , bacterial counts were significant reduced to 20–30% in samples treated with pro-coagulant MVs , PLs , or tissue factor , when compared with ctrl . MVs . Samples incubated with buffer only , served as a control ( 100% growth ) . Clot formation appears to be the critical moment in these experiments , since no reduction in bacterial growth was monitored when calcium was omitted and thus clotting prevented ( Figure 5B ) . Similar results , though not a complete reversion , were seen when recalcified samples were treated with a peptide ( Gly-Pro-Arg-Pro ) that prevents the polymerization of fibrin monomers ( Figure 5C ) [21] . To visualize the bacteria , samples were subjected to scanning electron microscopy ( Figure 5D–F ) . In the absence of pro-coagulant MVs , S . pyogenes bacteria appear as intact cocci when incubated for 30 min in plasma ( Figure 5D ) . In the next series of experiments , pro-coagulant MVs ( Figure 5E ) were added to the plasma bacteria mixture . Figure 5F illustrates that after activation with pro-coagulant MVs , bacteria were weaved in a fibrin network . It also appears that the morphology of bacteria was not significant compromised , as the bacterial cell membrane seems to be still intact ( Figure 5F ) . These images may indicate that the effect seen is of bacteriostatic nature rather than bactericidal , however , more experimental support is needed to prove this conclusion . Taken together the data suggest that pro-coagulant MVs are able to prevent bacterial spreading and impair bacterial proliferation inside a plasma clot . Recently we reported that pro-coagulant MVs from patients suffering from streptococcal sepsis are significant increased [13] . To test whether this can also be observed in an invasive animal model of streptococcal infection , mice were subcutaneously infected . The animals were sacrificed after three time points ( 10 , 24–30 , and 42–48 hours after infection ) and plasma samples were recovered by cardiac puncture . Figure 6A depicts that the TF content in the plasma samples was not significant raised 10 hours after infection , but was significantly increased at the later time points ( Figure 6A ) . Similar results were seen when measuring the concentrations of pro-coagulant MVs , though they already start to peak 10 hours after infection ( Figure 6B ) . Thus , the data show that the generation of pro-coagulant MVs is part of the host response to invasive infection with S . pyogenes . The role of MVs in systemic infectious diseases is currently not completely understood , but it has been speculated that elevated levels in the early phase of sepsis may have protective effects [22] . We therefore studied whether the local application of pro-coagulant MVs to the site of infection may improve the outcome of the disease . Three groups of mice were infected with S . pyogenes bacteria and were treated either with vehicle , ctrl . MVs or pro-coagulant MVs . While application of ctrl . MVs failed to improve survival as compared to control ( vehicle ) , treatment with pro-coagulant MVs significant prolonged survival time and decreased the mortality rate ( Figure 7A ) . The subcutaneous injection of pro-coagulant MVs also had an impact on the bacterial load in different organs of the infected mice . Mice received a subcutaneous injection of S . pyogenes bacteria and simultaneously a single dose of pro-coagulant MVs . Infected animals were sacrificed 18 hours after infection , and bacterial loads in the blood , liver , and spleen were determined . As depicted in figure 7B–D , treatment with pro-coagulant MVs resulted in decreased numbers of bacteria in all organs when compared with non-treated animals . These results are in line with previous conclusions and may indicate that pro-coagulant MVs are part of the early host defense to an infection at an early stage of the infectious disease progression .
Pro-coagulant MVs constitute one of the main reservoirs of blood-borne TF , which are released from monocytes , macrophages , or endothelial cells with inducible TF expression [23] and they are therefore considered to be key determinants of the hemostasis equilibration [24] . Notably , the number of pro-coagulant MVs can significantly increase in patients suffering from sepsis as reported by us and other groups [13] , [25] . These findings raise the question whether they are part of the host response to infection or rather contribute to systemic hemorrhagic complications , such as disseminated intravascular coagulation ( DIC ) in severely ill patients . Reid and Webster recently published a review article on the role of MVs in sepsis [22] . The authors conclude that MVs are beneficial at the early stage of sepsis as they can compensate for some of the host's systemic reactions [22] . Our findings support this notion , because local treatment with pro-coagulant MVs significantly prevents bacterial dissemination and improves survival . Moreover , activation of PMBCs is triggered by the binding of M1 protein to toll-like receptor 2 [12] , which suggests that formation and release of pro-coagulant MVs follows the principles of pattern recognition and are therefore part of the innate immune reaction . However , as seen for many other host defense mechanisms , the systemic induction of pro-coagulant MVs may contribute to severe complications , such as DIC . A better understanding of the molecular mechanisms that modulates the tightly regulated process may lead to the development of novel antimicrobial therapies with different modes of action that can be used for local treatment or in systemic complications . Microvascular thrombosis and the formation of a fibrin network can be considered as an efficient and early response of the host defense against bacteria that can lead to an immobilization of bacteria and thereby attenuates the spreading of the pathogen [19] , [20] , [26] . Our studies show that pro-coagulant MVs bind to S . pyogenes and that this interaction leads to an alteration of the bacterial surface into a pro-coagulative state . We found that fibrinogen is a docking molecule that attaches protein M1 , a streptococcal surface-bound adhesion factor , to pro-coagulant MVs . Subsequent mass spectroscopic analysis revealed an up-regulation of the fibrinogen binding integrins ( CD18 and CD11b , respectively ) at the surface of pro-coagulant MVs . This chain of events presents a plausible explanation as to how pro-coagulative MVs achieve their affinity for S . pyogenes . However , it cannot be ruled out that other proteins such as fibronectin , vitronectin or laminin are also involved [27]–[29] . Notably , many of these host adhesion factors are also receptors for other bacterial pathogens [30] . Thus their binding to pro-coagulant MVs may represent some kind of pattern recognition mechanism that allows the targeting of other microorganisms in a more general sense . Future work will show whether pro-coagulant MVs are also interacting with other bacterial species and whether this involves the recruitment of fibrinogen and/or other host adhesion proteins . Our results show that plasma clots that were induced by pro-coagulant MVs can immobilize S . pyogenes as efficiently as clots induced by tissue factor or artificial phospholipids . Importantly , clots formed in the presence of pro-coagulant MVs had antimicrobial activity against S . pyogenes , which could be explained by their cargo containing antimicrobial peptides and proteins . However , we noted that clots formed by the addition of tissue factor or artificial phospholipids , were also able to kill the entrapped bacteria . It therefore remains to be determined as to what extent the peptides/proteins with antimicrobial activity from pro-coagulant MVs contribute to bacterial killing , or if there are bactericidal substances generated during the activation of the coagulation cascade . The latter hypothesis is supported by recent findings that many coagulation factors contain a sequence at their carboxy-terminal part with an antimicrobial activity [31] , [32] . Taken together , our data show that activation of the coagulation cascade and the formation of a fibrin network are important mechanisms to prevent bacterial dissemination and proliferation . As pro-coagulant MVs are induced at an early stage during bacterial infection , their local interaction with bacteria can be considered as part of the early immune response .
The S . pyogenes strain AP1 ( 40/58 ) serotype M1 and its M1-derivate MC25 have been described previously [18] , [33] . All other S . pyogenes strains were clinical isolates from our strain collection that have been characterized by standard microbiological procedures . Bacteria were grown overnight in Todd-Hewitt broth ( THB; GIBCO ) at 37°C and 5% CO2 . M1 protein was purified from the supernatant of S . pyogenes MC25 , as previously described [18] . Artificial pro-coagulant phospholipids were from Rossix ( Sweden ) . Recombinant tissue factor and anti TF were from American Diagnostica ( Germany ) . Anti CD14 was from Dako ( Denmark ) . PBMC isolation , stimulation as well as MV purification were performed as described previously and used at concentration range from 50 to 150 MPs/µl [13] . S . pyogenes bacteria from 10 ml overnight culture were washed and resuspended in 1 ml 10 mM HEPES-buffer ( 2×109 CFU/ml ) . 150 µl bacteria and 30 µl MVs - in the presence or absence of 300 µl human plasma – were mixed and incubated for 30 min . at 37°C . Alternatively , fibrinogen depleted plasma ( Affinity Biologicals , Canada ) was used . Bacteria were washed 3 times with HEPES-buffer by centrifugation ( 1550× g for 10 min . ) , and finally resolved in HEPES-buffer . Clotting time was measured in a coagulometer ( Amelung ) after addition of reaction mixtures to recalcified normal human plasma . 100 µl recalcified normal plasma was mixed with 25 µl 2×105 CFU S . pyogenes bacteria , 25 µl MVs or pro-coagulant PLs ( 0 . 25 mM , Rossix ) , or 2 pM tissue factor ( American Diagnostica ) and incubated at 37°C for 5 min . The clots were covered with 10 mM Tris-buffer containing 1% plasma . At the indicated time points , 100 µl aliquots of the supernatant were plated onto blood agar in 10-fold serial dilutions and the number of bacteria was determined by counting colonies after 18 hours of incubation at 37°C . Plasma clots were produced as described above , covered with Tris-buffer containing 1% plasma and incubated at 37°C for 30 min . Alternatively , the tetrapeptide Gly-Pro-Arg-Pro ( Bachem ) was added to prevent clotting ( 1 . 5 mg/mL final concentration ) . After incubation clots were disrupted in a Ribolyser ( Hybaid , 30 sec at speed 4 . 0 ) and the homogenate was plated directly onto blood agar . The number of bacteria was determined by counting colonies after 18 hours of incubation at 37°C . MVs were labeled with the red fluorescence aliphatic chromophore PKH26 dye ( Sigma ) , which intercalate into lipid bilayers [34] . After labeling , MVs were washed and centrifuged as described [13] . 150 µl bacteria ( 2×109 CFU/ml ) and 30 µl labeled MVs were mixed in 300 µl human plasma and incubated for 30 min at 37°C . After incubation 10 µl of the mix was dropped onto a cover slide , counterstained with DAPI ( Invitrogen ) and visualised by a BX60 fluorescence microscope and 100×1 . 3 or 60×1 . 25 UplanFl objectives ( Olympus , Hamburg , Germany ) . Human proteins ( annexin V , anti CD14 AB , and anti TF AB ) were labeled with colloidal gold ( 15 and 5 nm in diameter , BBI International ) as described earlier [35] . MV/S . pyogenes preparations were mixed with gold-labeled 20 nM proteins for 20 min at room temperature and processed for negative staining [36] . Clots were fixed with 2 . 5% glutaraldehyde overnight . Samples were washed 2–3 times with 0 . 1 M sodium phosphate buffer ( pH 7 . 3 ) , dehydrated with a series of increasing ethanol concentrations ( 5 minutes in 30% , 5 minutes in 50% , 10 minutes in 70% , 10 minutes in 90% and two times 10 minutes in ethanol absolute ) , and dried with CO2 by critical point method with a Emitech dryer as outlined by the manufacturer . Dried samples were covered with gold to a 10 nm layer and scanned with a Zeiss DSM 960A electron microscope . The Actichrome TF activity assay kit ( American Diagnostica ) was used to quantify the TF pro-coagulant activity in the plasma samples [13] . The Coa-MP activity kit ( Coachrom Diagnostica ) was used according to the instructions of the manufactory , to measure the pro-coagulant activity of MVs in plasma [13] . Protein digestion was carried out as previously described [37] . The resulting peptide mixtures were concentrated using spin-columns from Harvard Apparatus using the manufactures' instructions . The hybrid Orbitrap-LTQ XL mass spectrometer ( Thermo Electron , Bremen , Germany ) was coupled online to a split-less Eksigent 2D NanoLC system ( Eksigent technologies , Dublin , CA , USA ) . Peptides were loaded with a constant flow rate of 10 µl/min onto a pre-column ( Zorbax 300SB-C18 5×0 . 3 mm , 5 µm , Agilent technologies , Wilmington , DE , USA ) and subsequently separated on a RP-LC analytical column ( Zorbax 300SB-C18 150 mm×75 µm , 3 . 5 µm , Agilent technologies ) with a flow rate of 350 nl/min . The peptides were eluted with a linear gradient from 95% solvent A ( 0 . 1% formic acid in water ) and 5% solvent B ( 0 . 1% formic acid in acetonitrile ) to 40% solvent B over 55 minutes . The mass spectrometer was operated in the data-dependent mode to automatically switch between Orbitrap-MS ( from m/z 400 to 2000 ) and LTQ-MS/MS acquisition . Four MS/MS spectra were acquired in the linear ion trap per each FT-MS scan which was acquired at 60 , 000 FWHM nominal resolution settings using the lock mass option ( m/z 445 . 120025 ) for internal calibration . The dynamic exclusion list was restricted to 500 entries using a repeat count of two with a repeat duration of 20 seconds and with a maximum retention period of 120 seconds . Precursor ion charge state screening was enabled to select for ions with at least two charges and rejecting ions with undetermined charge state . The normalized collision energy was set to 30% , and one microscan was acquired for each spectrum . The data analysis was performed as previously described [37] . Briefly , the MS2 spectra were searched through the X ! Tandem 2008-05-26 search engine [38] against the human protein database . The search was performed with semi-tryptic cleavage , specificity , 1 missed cleavages , mass tolerance of 25 ppm for the precursor ions and 0 . 5 Da for fragment ions , methionine oxidation as variable modification and cysteine carbamidomethylation as fixed modification . The database search results were further processed using the Trans-Proteomic pipeline , version 4 . 4 . 0 [39] . Real time biomolecular interaction was analyzed with a BIAcore3000 system ( Biosensor , La Jolla , CA ) using L1 sensor chips [40] . The L1 sensor chip comprises a carboxymethyl dextran hydrogel derivatized with lipophilic alkyl chain anchors to capture phospholipid vesicles . Experiments were performed at 25°C with 10 mM TRIS , 0 . 9% NaCl , pH 7 . 4 as running buffer and PBS ( pH 7 . 4 ) as immobilization buffer . MVs were coated onto the L1-sensor chip according to the manufacturer' s instructions . Briefly , the L1 chip surface was washed by 2×3 minute injections of 40 mM N-octyl-β-D-glucopyranoside ( Roth , Germany ) at a flow-rate of 10 µl/min . MVs in PBS were then injected over the sensor for 30 min at a flow-rate of 2 µl/min resulting in 2000–2200 response units ( RUs ) of ctrl or pro-coagulant MVs . To remove residual multilayer structures and loosely bound vesicles , a short pulse of 10 mM NaOH was applied . Subsequently , BSA ( 0 . 1 mg/ml , 5 min ) was added to block non-specific surface binding . The resulting bilayer linked to the chip surface was taken as a model MV-membrane surface for studying fibrinogen binding . Pure buffer solutions and the solution containing fibrinogen or M1 protein were applied at a flow rate of 10 µl/min . Following each cycle of analysis , the sensorchip was regenerated either with short pulses of 10 mM NaOH leaving the MV-lipid monolayer intact for additional interaction studies , or with 40 mM N-octyl-β-D-glucopyranoside stripping the MV-lipid layer from the surface in order to adsorb new MVs . The sensorgrams for each fibrinogen–MV bilayer interaction were analyzed by curve-fitting . Data from 5 concentrations were selected for statistic analysis . RUs of the running buffer were subtracted from the RUs of the sample solution . The data were analyzed using the Biaevaluation 3 . 0 software ( Biacore ) that offers various reaction models to perform complete kinetic analyses . The data from the BIAcore sensorgrams were fitted globally , and the heterogeneous ligand model resulted in optimum mathematical fits , reflected by low χ2 values ( <5 ) . The subcutaneous infection model with S . pyogenes AP1 strain were performed in female Balb/C mice as described previously [33] . This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Committee on the Ethics of Animal Experiments the Landesveterinär- und Lebensmitteluntersuchungsamt Rostock ( Permit Number: 7221 . 3-1 . 1-031/10 . ) . Statistical analysis was performed using GraphPad Prism , Version 4 . 00 . The P-value was determined by using the unpaired t-test ( comparison of 2 groups ) or the log-rank test ( comparison of survival curves ) . All samples were analyzed in triplicate and all experiments were performed at least three times , if not otherwise declared . The bars in the figures indicate standard deviation . | The coagulation system is much more than a passive bystander in our defense against exogenous microorganisms . Over the last years there has been a growing body of evidence pointing to an integral part of coagulation in innate immunity and a special focus has been on bacterial entrapment in a fibrin network . However , thus far , pro-coagulant MVs have not been discussed in this context , though it is known that their numbers can dramatically increase in many pathological conditions , including severe infectious diseases . In the present study we see a significant increase of pro-coagulant MVs in an invasive streptococcal mouse model , suggesting that their release is an immune response to the infection . We find that pro-coagulant MVs bind to Streptococcus pyogenes and promote clotting , entrapment , and killing of the bacteria in a fibrin network . As a proof of concept pro-coagulant MVs were applied as local treatment in the streptococcal infection model , and it was demonstrated that this led to a significantly improved survival in mice . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] | [
"medicine",
"biology"
] | 2013 | A Novel Role for Pro-Coagulant Microvesicles in the Early Host Defense against Streptococcus pyogenes |
Constraints in embryonic development are thought to bias the direction of evolution by making some changes less likely , and others more likely , depending on their consequences on ontogeny . Here , we characterize the constraints acting on genome evolution in vertebrates . We used gene expression data from two vertebrates: zebrafish , using a microarray experiment spanning 14 stages of development , and mouse , using EST counts for 26 stages of development . We show that , in both species , genes expressed early in development ( 1 ) have a more dramatic effect of knock-out or mutation and ( 2 ) are more likely to revert to single copy after whole genome duplication , relative to genes expressed late . This supports high constraints on early stages of vertebrate development , making them less open to innovations ( gene gain or gene loss ) . Results are robust to different sources of data—gene expression from microarrays , ESTs , or in situ hybridizations; and mutants from directed KO , transgenic insertions , point mutations , or morpholinos . We determine the pattern of these constraints , which differs from the model used to describe vertebrate morphological conservation ( “hourglass” model ) . While morphological constraints reach a maximum at mid-development ( the “phylotypic” stage ) , genomic constraints appear to decrease in a monotonous manner over developmental time .
To what extent do the processes of embryonic development constrain genome evolution ? Correlations between developmental timing and morphological divergence have long been observed , but the mechanisms and molecular basis of such patterns are poorly understood . The most commonly used measure of selective pressure on the genome , the ratio of non-synonymous to synonymous substitutions ( dN/dS ) in protein coding genes , has been of limited help in this case . Stronger constraints have been found on genes expressed in late embryonic stages in Drosophila [1] , but most other studies have failed to report robust evidence for a lower dN/dS ratio in genes expressed at constrained developmental stages [2]–[5] . A different approach has been to characterize which genes are duplicated , and which are not: studies of C . elegans [2] and Drosophila [6] have found less duplication of genes expressed in early development . These results show that it is possible to identify developmental constraints at the genomic level . They have a few limitations though . One is that the data available has limited the characterization of developmental time to broad categories such as “early” and “late” . A second is the difficulty of relating results from two derived invertebrate species , to morphological evolution models in vertebrates [7] . Indeed it is in vertebrates that the fundamental models of developmental constraint on evolution have been established , starting in the nineteenth century with the “laws” of von Baer [8] , claiming a progressive divergence of morphological similarities between vertebrate embryos , with the formation of more general characters before species-specific characters . Integration of these observations within evolutionary biology has not always been straight-forward [9]–[11] . More recently , an “hourglass” model was proposed to describe morphological evolution across development [12] , [13]: in the earliest stages of development ( cleavage , blastula ) there is in fact a great variety of forms in vertebrate embryos . Later in development , a “phylotypic” or conserved stage is observed , where many morphological characteristics are shared among vertebrates . This stage is usually presumed to be around the pharyngula stage . After this bottleneck , a “von Baer-like” progressive divergence is again observed . The conserved phylotypic stage has been explained by assuming higher developmental constraints [13]–[15] . The limits on morphological evolution would be placed by the structure of animal development , making some changes unlikely or impossible . How such limitations are encoded in the genome , or impact its evolution , is still an open question . In this work , we investigate the existence and timing of constraints on genes expressed in vertebrate development . We use representatives of the two main lineages of vertebrates , a teleost fish and a tetrapode , and we explore the impact of experimental gene loss , and of gain of gene copies in evolution . We find that timing of development has a strong impact in both cases , but that the pattern of constraints on genome evolution does not follow the morphological hourglass model . High constraints are present in early stages of development and relax progressively over time .
First , we used the phenotypes of gene loss-of-function as an indicator of selective pressure on genes . We extracted genes essential for the viability of the zebrafish , giving a lethal phenotype when non functional [16] . We expect that the loss of a gene should be more deleterious if this gene is expressed at a developmental stage with strong constraints . Thus we estimated whether genes were expressed or not at each stage , and computed the ratio of expressed essential genes to expressed reference genes ( no reported loss of function phenotype ) . We then plotted the variation across development of this ratio . We used two different types of data to evaluate the presence of gene expression: ( i ) expression patterns from in situ hybridizations ( Figure 1A ) , and ( ii ) “present” or “absent” calls from an Affymetrix microarray experiment ( Figure 1B ) . Results are consistent for both data types: the proportion of essential genes is higher among genes expressed in early development , with a significant negative correlation . For the in situ hybridizations ( Figure 1A ) , a linear regression is significant , but a parabola is not . The parabola has been suggested as the quantitative expectation of an hourglass-like model 3 , 17 . These results indicate a continuous trend over developmental time , with stronger constraints on early development . Considering gene expression either “present” or “absent” allows straightforward statistical analysis , but it is a strong approximation of the continuous nature of gene expression . To take advantage of the quantitative signal from the microarray data , we contrasted the median expression level of all the essential genes to that of all of the reference genes ( Figure 2A ) . We used the median because it is less sensitive to extreme values [18]; results were consistent using the mean ( not shown ) . To estimate the significance of the difference between the two curves , we performed a randomization test ( see Methods ) , which provides 1% and 1‰ confidence intervals ( Figure 2B ) . The expectation is now that the essential genes should be enriched in genes highly expressed at the stages with strong constraints . And consistently with the previous observations , essential genes are significantly more expressed in early stages ( until 11 . 7 hours ) , and less expressed in late stages of development ( from 5 days to 14 days ) . No specific trend is visible around the phylotypic stage . Similar results are obtained for genes which give an “abnormal” phenotype after loss of function ( Text S1 and Figure S4 ) . To complement this approach , we defined groups of genes according to their expression pattern during development ( see Methods ) . This clustering of zebrafish genes provided us notably with a cluster of 2446 genes with high expression in early development , decreasing over time ( Figure 3 , cluster 3 ) , and an opposite cluster of 1123 genes lowly expressed in early development , increasing over time ( Figure 3 , cluster 4 ) . As expected , genes whose expression is highest in early development are more frequently essential ( 1 . 1% vs . 0 . 6% ) , and induce more frequently abnormal phenotypes when non functional ( 6 . 1% vs . 2 . 9% ) . We performed a similar analysis in mouse , with some differences of methodology due to the data available . For expression , we used of a large amount of EST ( Expressed Sequence Tags ) data from libraries spanning development , from which we deduced presence or absence of expression ( see Methods ) . Only phenotypes obtained by the targeted knock-out technique were used . As knock-out experiments with no observable phenotype are reported in mouse , we can use these as a reference set , instead of non annotated genes as in zebrafish . The ratio of expressed essential genes to expressed reference genes is significantly negatively correlated with developmental time ( Figure 4A ) , as in zebrafish ( Figure 1 ) . Repeating the same approach with genes inducing a phenotype reported as “abnormal” when they are not functional , no significant trend is detected compared to genes inducing no phenotype , after multiple testing correction ( Figure 4B ) . Moreover , these genes can be used as a reference for essential genes ( Figure 4C ) , with results very similar to the use of genes inducing no phenotype after loss of function ( Figure 4A ) . Thus in mouse , genes inducing abnormal phenotypes when non-functional have a behavior more similar to the reference set of “non essential” genes . The fish specific whole genome duplication [19] provides us with a natural experiment on constraints on gene doubling: after this event approximately 85% of duplicated genes lost one copy , and the subset which retained both copies is known to be biased relative to function and selective pressure [20] . Thus we tested if duplicate gene expression pattern in zebrafish development was biased compared to singletons . We plotted the median expression profiles of duplicates originating from the fish specific whole genome duplication , and of singletons , genes whose duplicate copy has been lost after the genome duplication ( Figure 5 ) . Duplicates are less expressed in early stages of development . The difference of median expression decreases progressively , similar to the observations for essential or abnormal phenotype genes . Larval time points show a maximum expression of duplicates relative to singletons . Two scenarios can explain this result . First , retention of two copies may be more likely after the whole genome duplication for genes less expressed in early development . Second , the retention of genes may be unbiased relative to development , but duplicate genes may evolve secondarily lower expression in early development . To get a proxy of the ancestral state before whole genome duplication , we used again mouse data , which has diverged from zebrafish before the fish specific duplication . We compared mouse orthologs of zebrafish duplicates to mouse orthologs of zebrafish singletons , regarding their expression in development ( Figure 6 ) . Mouse orthologs of duplicates are significantly less expressed in early development compared to orthologs of singletons . This result in mouse is consistent with the observations in zebrafish , and the most parsimonious explanation is that expression was similar in the ancestor of the two lineages . Therefore we can accept the first hypothesis: after the fish specific whole genome duplication , there was preferential retention of duplicates less expressed in early development . To check if this phenomenon is particular to the fish specific genome duplication , we repeated this analysis with the two ancient rounds of genome duplication ( “2R” ) , which occurred in the ancestor of vertebrates [21] . It is difficult to distinguish between the two whole genome duplications since no model species diverged from the vertebrate lineage between them . Therefore we looked at the median expression profiles of genes with any duplication at the origin of vertebrates , compared to singletons , whose duplicates were lost after both whole genome duplications . For zebrafish , we restricted this analysis to genes which are singletons regarding the fish specific whole genome duplication . Similarly to fish specific duplicates , duplicates from 2R are significantly less expressed than singletons in the early development of zebrafish ( Figure S1 ) and mouse ( Figure S2 ) . Thus mechanisms of retention after whole genome duplication seem to be conserved during vertebrate evolution ( see also Text S1 ) . To check if sequences of genes expressed at different stages in development are experiencing different selective pressure , we used the non synonymous to synonymous substitution ratios ( dN/dS ) . In zebrafish , we used an approach similar to Davis et al . [1]: at each stage we performed the correlation between dN/dS and gene expression from microarray data ( Figure S3 ) . It has been shown that genes retained in duplicate tend to evolve slowly [20] , [22] . To control for that factor , we kept only strict singletons in the analysis ( genes whose duplicate was lost after 2R and fish-specific genome duplications ) . At all stages the correlation is negative , confirming that genes with higher expression levels are under stronger purifying selection [23] , [24] . We note that correlation at the “adult” stage ( 90 days ) is weaker ( Figure S3 ) : the link between expression and selective constraints on sequences appears stronger in development than in adult . But there is not a significant trend over time ( Spearman ρ = 0 . 08; p = 0 . 68 ) . In mouse , we considered only singletons after 2R genome duplication , and we compared the slowest evolving genes ( 25% lower dN/dS ) with the fastest evolving genes ( 25% higher dN/dS ) . There is a significant correlation with time of expression ( Figure 7 ) . Genes with strong sequence constraints ( low dN/dS ) tend to be expressed early in development . What is the function of the genes whose evolution is constrained by expression in early development ? We analyzed enrichment or depletion in Gene Ontology [25] categories for the clusters based on gene expression ( Figure 3 ) . Using the Molecular Function ontology , genes whose expression is highest in early development are significantly enriched in fundamental processes of the cell , such as RNA processing , transcription , and DNA replication ( Table S1 ) . This is very similar to the categories observed to be enriched in house keeping genes [26] . It is also consistent with the categories depleted in fish specific duplicates [20] . Conversely , genes highly expressed in early development are depleted in receptor or channel activity , while these activities are enriched in genes highly expressed in late development . Fewer terms are significant for the Biological Process ontology , and results are essentially consistent with the Molecular Function . Overall , the genes expressed in early development , which appear constrained against gene duplication or loss of function , seem to be house keeping genes involved in basic cellular processes .
Recent discussion of the evolution of ontogeny [27] has allowed the clarification of several important points . The first is that models must be explicitly defined , to allow testing . Poe and Wake [17] distinguish three models for the evolution of ontogeny: the early conservation model à la von Baer [8]; the hourglass model , characterized by a conserved phylotypic stage [12] , [28]; and the adaptive penetrance model ( an inverted hourglass ) . The second point is that quantitative testing is important to distinguish between these models . At the morphological level , several studies have used heterochrony data from vertebrates to quantify the amount of change at each stage of development [17] , [29] . Surprisingly , this led to rejection of both the early conservation and the hourglass models , although which model is favoured remains disputed [27] . The third point that should be clarified is the distinction between constraints at the level of patterns , and constraints at the level of processes [29] . The studies of heterochrony in vertebrates are typically concerned with the pattern . In this framework , our results clearly provide a quantitative test which supports the early conservation model . By studying not morphological structures but features of the genome and its expression , this test concerns the level of processes , not patterns . Thus an important point to be made is that our results should be taken neither in contradiction nor in support of any specific model at the level of patterns , given our still limited knowledge of causal relationships between process and patterns in ontogeny [30] . On the other hand , our results do appear to be in contradiction with previous reports of a maximum of constraints on processes around the phylotypic stage of vertebrates [3] , [4] , [31] . We use two simple measures of constraint on the expression of a gene at a developmental stage: if expression of one copy is needed , then ( i ) removing it may be deleterious , and ( ii ) increasing the number of copies may also be deleterious . This view is consistent with a recent study in yeast which suggests that constraints influencing the ability to lose certain genes or to maintain them in duplicate may be similar [32] . We expect gain or loss of genes highly expressed at more constrained developmental stages to be counter-selected . And indeed , we find a clear and significant trend: early development is strongly constrained , then constraints diminish during development in a continuous manner . Genes highly expressed in early development are more frequently essential , and less frequently preserved in double copy after genome duplication . Thus early development is less robust against gene loss and against gene doubling . Trends are conserved between mouse and zebrafish , representatives of the two main lineages of bony vertebrates , and between 2R and fish specific genome duplications . An indication of how strong these constraints are is our capacity to predict which genes were kept in duplicate in zebrafish based on expression pattern in mouse . Despite more than 400 MY of independent evolution , and the use of relatively noisy data ( mix of EST libraries ) , more than a quarter of the variance in gene retention is explained ( Figure 6; r2 = 0 . 27 ) . There is also some signal for early conservation at the level of coding sequences , at least in mouse ( Figure 7 ) . What we do not see is any genomic evidence for specific constraints at a phylotypic stage . Both in zebrafish and in mouse , the pharyngula stage appears to be part of the general trend from stronger genomic constraints in early development , towards weaker genomic constraints at later stages . We believe that our data are sufficiently detailed , and exhibit sufficiently strong signal , that a maximum of genomic constraints at the phylotypic stage would be visible . So where does the contradiction with previous studies come from ? An early quantitative study [31] found that when screens were done in rodents for the induction of teratogenesis , most abnormalities were obtained by applying teratogens during the phylotypic stage . This was interpreted [31] as supporting strong constraints at the phylotypic stage , due to inductive interactions . But these screens aimed not to test developmental robustness , but to obtain abnormal embryos for experimental work . As remarked by Bininda-Emonds et al . [29] , Galis and Metz [31] define the phylotypic stage broadly as including most organogenesis . If application of teratogens in early development resulted in lethality before organogenesis , it would not be of interest to the researchers performing the screens . Thus it seems that what Galis and Metz [31] measured was the potential for a stage to produce morphological abnormalities , not the overall constraints on ontogeny at each stage . There seems to be little reason to suppose that such data provide “an accurate model of natural selection” [33] , unlike e . g . the retention of duplicate genes over long evolutionary periods . It is worth noting that we observe a “peak” of constraints shortly after pharyngula ( Figure 4B ) for the expression profile of mouse genes which give an “abnormal” phenotype when knocked-out . The behavior of these genes is surprising , because in zebrafish the trend for such genes was similar to that for essential genes . We suspect that the definition of abnormal phenotypes differs between databases and between investigators working in different species . Less severe phenotypes may be reported as “abnormal” in mouse , relative to zebrafish . Of note , data in ZFIN [16] come mainly from the reviewed literature , where minor abnormalities of phenotype are rarely reported , whereas data in the MGD [34] come also from genome wide mutagenesis , and thus include such minor abnormalities . Minor abnormalities in mouse phenotype may also be easier to detect because of the gross similarity with human in anatomy and physiology . In any case , these are the data in our study which most closely approximate the teratogenesis study , and the only data that do not support the early conservation model . Although this trend is statistically not significant , it is consistent with the observations of Galis and Metz [31] . This deserves to be further examined in future studies . Two other studies which quantified a maximum of constraints at the phylotypic stage did use evolutionary measures of constraint . These studies [3] , [4] estimated constraints on the evolution of coding sequences , in relation to the timing of expression in mouse development from EST data . Despite similar experimental designs and data , we reached differing conclusions . First , we note that we did check for sequence conservation ( dN/dS ) trends over development . In zebrafish , we found no robust pattern ( Figure S3 ) , while in mouse we found support for the early conservation model ( Figure 7 ) . Second , in our analyses we found that small samples of ESTs could introduce important variability , which is why we used weighted regressions for all computations based on these data . For example , we see a very high ratio of mouse orthologs of zebrafish singletons to duplicates for Theiler stage 5 ( day 4 ) ( Figure 6 ) ; but this is obtained based on only 628 genes with at least one EST at that stage ( median over all stages: 3767 ) . The weighted regression insures that such a point has a weak incidence on the statistical significance . Similar issues are visible in the data of Irie et al . [4] , but are not addressed in their analysis . Indeed , the extreme points they use to support constraints at pharyngula are based on some of the smallest samples of their dataset . Finally , it should be noted that another study in mouse found an opposite pattern ( relaxation of constraints near the phylotypic stage ) using an alternative measure of constraints on sequences , the ratio of radical to conservative amino acid changes , KR/KC [5] . In our opinion , these contradictory and weakly supported results are consistent with the idea that overall , coding sequence change seems to have a rather modest contribution to the evolution of development . This is consistent with a stronger contribution of regulation of expression [35] , [36] . Our results were obtained on data which either reflect the action of natural selection ( duplicate gene retention ) , or are directly relevant to fitness ( loss-of-function lethality ) , and provide unambiguous trends with strong statistical support . Moreover , the consistent patterns in zebrafish in situ hybridization and microarray data , and mouse EST data , show robustness to potential experimental biases or sampling errors . The early conservation model for genomic processes is reinforced by the enrichment of early expressed genes in fundamental cellular processes ( Figure 3; Table S1 ) . This is the opposite of duplicated genes , which may be more involved in innovation , and have been reported to be enriched in developmental or behavioural processes [20] , [21] . Our results are consistent with the observation that basic cores of gene regulatory networks ( GRNs ) are highly constrained in early stages of animal development [37] , [38] , although we add the notion of a progressive decrease in constraints . This indicates that some relations between the timing of cell-fate decisions in development and rates of genome evolution may be widely shared among animals [7] , [39] . Indeed , many studies underline gastrulation as a crucial step in development [40] , [41] . Accordingly this period is shown here to be subject to highest constraints , consistent with the famous Lewis Wolpert quote: “It is not birth , marriage , or death , but gastrulation , which is truly the most important time in your life” [42] .
Microarray data of zebrafish ( Danio rerio ) development were downloaded from ArrayExpress ( E-TABM-33 ) [43] . This experiment uses an Affymetrix GeneChip Zebrafish Genome Array ( A-AFFY-38 ) . 15 stages were sampled , spanning from fertilization to adult stages ( 15 minutes , 6 , 8 , 9 , 10 , 11 . 7 , 16 , 24 , 30 hours , 2 , 4 , 5 , 14 , 30 , 90 days , covering zygote , segmentation , gastrula , pharyngula , hatching , larval , juvenile , adult ) . Two replicates were made per time point; we use both of them for computations , and the 2 values are plotted to give an order of the variability between replicates . Raw CEL files were renormalized using the package gcRMA [44] of Bioconductor version 2 . 2 [45] . We used the “affinities” model of gcRMA , which uses mismatch probes as negative control probes to estimate the non-specific binding of probe sequences . The normalized values of expression are in log2 scale , which attenuates the effect of outliers . Mapping of D . rerio genes on Affymetrix probesets was made using Ensembl [46] annotation for zebrafish genome version Zv7 ( unpublished ) . We did not consider the first time point of the data ( 15 minutes , fertilization ) . Its behaviour was peculiar in many cases . We explain this by the presence of maternal transcripts in the embryo [47] . These transcripts are largely degraded by 6 hours of development [48] , the second time point of the dataset . For the absolute detection of transcripts ( presence or absence calls ) , the method we used [49] replaces all MM probe values by a threshold value which is based on the mean PM value ( after gcRMA transformation ) of probesets that are very likely to have absent target transcripts . This removes the influence of probe sequence affinity and results in better performance than the MAS 5 algorithm . For the zebrafish microarray data we first used a randomization approach to assess the significance of the difference between two curves of median expression across development ( for example median expression of duplicates vs . singletons , or of essential genes vs . genes with no reported phenotype ) . If the two groups contain n1 and n2 genes , we pooled all these genes and randomly separated them into two new groups of same sizes ( n1 and n2 ) . Then we calculated and recorded the difference between the two new curves of median expressions across development . After repeating this randomization 10 , 000 times , we could define 1‰ and 1% confidence intervals . Second , we calculated the Spearman correlation between developmental time and the difference between two curves of median expression across development . Bonferroni correction was applied to correct for multiple testing , considering the 9 tests computed with this microarray data ( Figure 1; Figure 2; Figure 5; Figure S1; Figure S3; Figure S5A–D ) : α = 0 . 05/9 = 0 . 0056 . In order to identify genes lowly or highly expressed in early development , we used the Fuzzy C-Means soft clustering method implemented in the Bioconductor package Mfuzz [50] . After a pre-filtering step ( genes with sd <0 . 5 were removed ) , we ran the algorithm with the number of clusters set to c = 4 . This gave one cluster of genes lowly expressed across development ( 3641 probesets , 2261 Ensembl genes ) , one of genes highly expressed ( 2175 probesets , 1175 Ensembl genes ) , one of genes whose expression increased ( 1714 probesets , 1123 Ensembl genes ) and one of genes whose expression decreased ( 3306 probesets , 2446 Ensembl genes ) ( Figure 3 ) . EST ( Expressed Sequence Tags ) data were retrieved from BGEE ( dataBase for Gene Expression Evolution , http://bgee . unil . ch/ ) , a database comparing transcriptome data between species [51] , including EST libraries from UniGene [52] . The mapping of UniGene clusters on Ensembl genes is taken from Ensembl ( version 48 ) [46] , where a percentage of identity of 90% is set as the minimum threshold to link an Ensembl gene with a UniGene cluster . Each library has been annotated manually to ontologies of anatomy and developmental stages , if it was obtained under non pathological conditions , with no treatment ( “normal” gene expression ) . We considered a gene expressed at one time point in development if at least one EST was mapped to this gene at this time point . Thus , we could retrieve the number of genes expressed at each time point of mouse ( Mus musculus ) development . From this set we extracted two groups to compare ( for example essential/non essential , or duplicates/singletons ) . As the total number of ESTs available at each time point is different , we use at each time point the ratio of the numbers of genes expressed in the two groups . We obtained similar results when we defined a gene as expressed if it had at least two ESTs mapped to it . Also , considering the ratio of the mean number of ESTs per gene at each stage , instead of the ratio of the number of genes expressed at each stage , gave similar results ( not shown ) . We used data from 297 EST libraries , spanning 26 different developmental stages ( from TS01 to TS27 ) , corresponding to a total of 633 , 307 ESTs . A weighted linear regression between developmental time and expression ratios was fit to the data , and a F-test was run to assess if the slope was significantly different from zero . Weights were the total number of genes expressed at each stage . Bonferroni correction was applied to correct for multiple testing , considering the 6 ratios tested with mouse EST data ( Figure 4A–C; Figure 6; Figure 7; Figure S2 ) : α = 0 . 05/6 = 0 . 0083 . To test for an hourglass-like model , we adjusted a parabola ( polynomial model of order 2 ) , as in Hazkani-Covo et al . [3] . We used an ANOVA to estimate if the increase in fit to the data ( r ) between the linear and parabola models was significant . The same Bonferroni correction was applied to the ANOVA . This test was never significant , providing no evidence for a maximum or a minimum of the ratio during development ( Dataset S2 ) . In situ hybridization expression data from ZFIN [16] were retrieved using BGEE [51] . We considered only stages with more than 1000 genes expressed , starting when maternal genes are largely degraded ( 6 hours post-fertilization [48] ) . We retrieved all genes with at least one report of expression by in situ hybridization , at each time point of zebrafish development . From this set we extracted two groups ( for example essential and non-annotated genes ) , and analyzed their ratio across development using the same methodology as with ESTs ( see above ) . The orthology relationships , and the values of dN ( rate of non-synonymous substitution per codon ) and dS ( rate of synonymous substitution per codon ) were obtained from Ensembl version 48 [46] . We retrieved zebrafish genes with one-to-one orthologs in Tetraodon nigroviridis and Takifugu rubripes ( divergence time is ∼32 MYA between the two pufferfish species and ∼150 MYA with Danio rerio [53] ) . We downloaded the pairwise dN and dS between Tetraodon and Takifugu , calculated with codeml from the PAML package in the Ensembl pipeline ( model = 0 , NSsites = 0 ) [54] . Ensembl considers that dS values are saturated when they reach a threshold which is 2*median ( dS ) . See http://www . ensembl . org/info/about/docs/compara/homology_method . html for further details . We selected a set of 4937 genes having dN , dS and Affymetrix expression data . Among them 620 genes were strict singletons in fishes ( loss of duplicates after 2R and after the fish-specific genome duplication ) . At each time point we performed the Spearman correlation between the dN/dS ratio and expression , following Davis et al . [1] . A t-statistic was used to assess if the correlation coefficient was different from 0 . For the analysis in mouse we retrieved pairwise dN and dS between human and mouse , for genes with one-to-one human orthologs ( 14 , 333 genes ) . We kept only the singletons for 2R genome duplication and separated the 25% with the highest dN/dS and the 25% with the lowest dN/dS ( 607 genes in each group ) . We then compared the expression across development of these two groups using EST data . Using the 10% highest and lowest dN/dS gave similar results ( not shown ) . Gene families were obtained from the HomolEns database version 3 ( http://pbil . univ-lyon1 . fr/databases/homolens . html ) , which is based on Ensembl release 41 [46] . HomolEns is build on the same model as Hovergen [56] , with genes organized in families , which include pre-calculated alignments and phylogenies . In HomolEns version 3 , alignments are computed with MUSCLE [57] ( with default parameters ) , and phylogenetic trees with PhyML [58] . Phylogenies are computed on conserved blocks of the alignments selected with GBLOCKS [59] . Using the TreePattern functionality of the FamFetch client for HomolEns , which allows scanning for gene tree topologies [60] , we selected sets of genes with or without duplications on specific branches of the vertebrate phylogenetic tree . Regarding the fish-specific whole genome duplication , we found 1772 Ensembl IDs for duplicates in zebrafish , 8821 for singletons in zebrafish , 755 mouse orthologs of these duplicates , and 6843 mouse orthologs of these singletons . For the 2R whole genome duplications , we found 986 duplicates and 1266 singletons in zebrafish , and 2448 duplicates and 2705 singletons in mouse ( Datasets S1 and S2 ) . Over and under representation of GO terms [25] was tested by means of a Fisher exact test , using the Bioconductor package topGO version 1 . 8 . 1 [61] . The reference set was all Ensembl genes mapped to a probeset of the zebrafish Affymetrix chip . The “elim” algorithm of topGO was used , allowing to decorrelate the graph structure of the gene ontology , reducing non-independence problems . A False Discovery Rate correction was applied , and gene ontology categories with a FDR <15% were reported . R was used for statistical analysis and plotting ( http://www . R-project . org/ ) [62] , in conjunction with Bioconductor packages ( http://www . bioconductor . org/ , version 2 . 2 ) [45] . To retrieve genomic information we used the BioMart tool [55] or connected to the Ensembl MySQL public database [46] . | Because embryonic development must proceed correctly for an animal to survive , changes in evolution are constrained according to their effects on development . Changes that disrupt development too dramatically are thus rare in evolution . While this has been long observed at the morphological level , it has been more difficult to characterize the impact of such constraints on the genome . In this study , we investigate the effect of gene expression over vertebrate developmental time ( from early to late development ) on two main features: the gravity of mutation effects ( i . e . , is removal of the gene lethal ? ) and the propensity of the gene to remain in double copy after a duplication . We see that both features are consistent , in both zebrafish and mouse , in indicating a strong effect of constraints , which are progressively weaker towards late development , in early development on the genome . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/functional",
"genomics",
"developmental",
"biology/developmental",
"evolution",
"evolutionary",
"biology/genomics",
"computational",
"biology/genomics",
"genetics",
"and",
"genomics/bioinformatics",
"evolutionary",
"biology/developmental",
"evolution"
] | 2008 | Developmental Constraints on Vertebrate Genome Evolution |
Increased incidence of hand , foot and mouth disease ( HFMD ) has been recognized as a critical challenge to communicable disease control and public health response . This study aimed to quantify the association between climate variation and notified cases of HFMD in selected cities of Shanxi Province , and to provide evidence for disease control and prevention . Meteorological variables and HFMD cases data in 4 major cities ( Datong , Taiyuan , Changzhi and Yuncheng ) of Shanxi province , China , were obtained from the China Meteorology Administration and China CDC respectively over the period 1 January 2009 to 31 December 2013 . Correlations analyses and Seasonal Autoregressive Integrated Moving Average ( SARIMA ) models were used to identify and quantify the relationship between the meteorological variables and HFMD . HFMD incidence varied seasonally with the majority of cases in the 4 cities occurring from May to July . Temperatures could play important roles in the incidence of HFMD in these regions . The SARIMA models indicate that a 1° C rise in average , maximum and minimum temperatures may lead to a similar relative increase in the number of cases in the 4 cities . The lag times for the effects of temperatures were identified in Taiyuan , Changzhi and Yuncheng . The numbers of cases were positively associated with average and minimum temperatures at a lag of 1 week in Taiyuan , Changzhi and Yuncheng , and with maximum temperature at a lag of 2 weeks in Yuncheng . Positive association between the temperature and HFMD has been identified from the 4 cities in Shanxi Province , although the role of weather variables on the transmission of HFMD varied in the 4 cities . Relevant prevention measures and public health action are required to reduce future risks of climate change with consideration of local climatic conditions .
Hand , foot and mouth disease ( HFMD ) is an emerging infectious disease mainly caused by highly contagious intestinal viruses human enterovirus 71 ( EV71 ) and coxsackievirus A16 ( Cox A16 ) [1–3] . It is a human syndrome characterized by a distinct clinical presentation of fever , accompanied by oral ulcers and maculopapular rash or vesicular sores on the hands and feet , and sometimes the buttocks . HFMD transmission is through close personal contact , exposure to feces , contaminated objects and surfaces of an infected person . In recent decades , HFMD has become a growing public health threat to children , particularly those under the age of 5 [4–6] . Epidemics of HFMD are frequent and widespread in Asian countries , especially in China , Singapore , Malaysia and Japan , which have documented many large outbreaks of HFMD with severe complications and deaths predominantly among children [2 , 7–12] . At present , there is no specific curative treatment , and vaccine development is still in progress [13] . Weather variables might play a certain role in the transmission of the disease , as time series analysis in Guangzhou and Shenzhen , in China , showed that weather variation could affect the disease occurrence with a short lag period [14 , 15] . Under the context of global environmental change , the frequency of HFMD epidemics may be projected to increase in the future due to continued viral mutation , climate change , and the lack of health resources and effective surveillance systems in some regions [14 , 16–18] . Therefore , risk detection , early warning of HFMD cases with the capacity to predict a possible epidemic , and efficient public health response will be important to minimize the risk of epidemics and adverse impacts of HFMD . Intestinal viruses have a worldwide distribution . In tropical and semitropical areas , they are present throughout the year , whereas in temperate climatic zones , they are more common during summer and fall [19] . A previous literature review has indicated that HFMD typically occurs in the summer and early autumn [20] . Such seasonal distribution suggests climatic variations may play a certain role in the transmission of the disease , particularly in temperate areas . The fifth assessment report of the Intergovernmental Panel on Climate Change ( IPCC AR5 ) pointed out that the globe has experienced surface warming and projections of annual average temperature changes for 2081–2100 under Representative Concentration Pathways ( RCPs ) 2 . 6 and 8 . 5 , relative to 1986–2005 [21] . In China , increases in temperature have been observed in most regions from south to north during the last decade [22] . Climate change has also been identified as an important risk factor for transmission of infectious diseases , especially vector and food borne diseases [23] . Since 2008 , the China Ministry of Health has listed HFMD as a notifiable Class-C communicable disease , which has been included in the national communicable disease surveillance system and reporting network . Over six million cases had been reported up to the end of 2012 in China [17] . Studies have examined the association between HFMD and climate variables in selected regions [14 , 24–26] . Meteorological parameters , such as temperature and relative humidity , may affect the transmission and the frequency of HFMD . However , the effects of climate variables are not consistent in published studies , which could be due to various local climatic conditions , socioeconomic status and demographic characteristics in different regions . In particular , an understanding of the impact of seasonality and meteorological variables on disease transmission remains limited . A comparison among different cities within a Province may minimize potential confounding effects of socioeconomic inequalities and demographic differences , which will be used in this study to examine the role of climate variation on the incidence of HFMD , using Seasonal Autoregressive Integrated Moving Average ( SARIMA ) models . The purpose of this study was to identify , with SARIMA models , the impact of meteorological variables on HFMD in 4 major cities of Shanxi in northern China , using existing surveillance data , and to quantify the relationship between climate variation and the incidence of HFMD . This study will provide scientific evidence to assist public health policy-making to carry out efficient prevention and control of HFMD .
The study was approved by the Ethics Committee of Shanxi Medical University ( No . 2013091 ) , China , and conducted in accordance with its guidelines . Shanxi HFMD data were provided by the Shanxi Center for Disease Control and Prevention and were obtained from the National Surveillance System . No informed consent was required because no individual-level analysis was performed . The information contained in the patients’ records was anonymized and de-identified prior to analysis . Only aggregated data were analyzed and reported . Shanxi Province , which is located in North China , has a temperate , continental , monsoonal climate with four distinct seasons . The average temperature in January is in the range -16°C to -2°C and in July between 19°C and 28°C , the average rainfall is between 350 to 700 mm , and the average daily sunshine is between 7 to 9 hours . This study selected 4 major cities from north to south ( Datong , Taiyuan , Changzhi and Yuncheng ) , which have similar socioeconomic and demographic conditions ( Fig . 1 ) . Datong ( latitude 40°2′30" N and longitude 113°35′50" E ) is the northernmost prefecture-level city of Shanxi Province , with a population of 3 . 36 million by the end of 2012 ( data from the Shanxi Bureau of Statistics ) . Taiyuan ( latitude 37°43' 36" N and longitude 112°28′14" E ) is the capital and largest city of Shanxi , which is located at the centre of the province with an East-West span of 144 km and a North-South span of 107 km , and a population of 4 . 26 million in 2012 . Changzhi ( latitude 36°11′0" N and longitude 113°6′0" E ) is the southeast city of Shanxi Province , with a population of 3 . 37 million in 2012 . Yuncheng ( latitude 35°1′33" N and longitude 111°0′19" E ) is a southwestern city in Shanxi , with a population of 5 . 19 million in 2012 . Daily meteorological data including precipitation , average temperature , maximum temperature , minimum temperature , average relative humidity , and hours of sunshine for the study period from 1 January 2009 to 31 December 2013 , were obtained from the Shanxi Meteorological Administration . Daily meteorological data were aggregated on a weekly basis which comprised a total period of 261 weeks . Being a notifiable disease [17] , all clinical and hospital doctors are required to report cases of HFMD to the local Center for Disease Control and Prevention . The diagnosis criteria for HFMD cases were provided in a guidebook published by the Chinese Ministry of Health [27 , 28] . Patients with HFMD have the following symptoms: fever , papules and herpetic lesions on the hands or feet , rashes on the buttocks or knees , inflammatory flushing around the rash and fluid in the blisters , or sparse herpetic lesions on the oral mucosa . A recent data quality survey report has demonstrated that the data are of high quality in China , with reporting completeness of 99 . 84% and accuracy of the information reported to be 92 . 76% [29] . In addition , in order to reduce apparent underreporting and a large number of missing information of patients from the early stage of the surveillance system in 2008 , only the data from 2009 to 2013 were used for analysis . The weekly data of HFMD cases for the period were obtained from the China Information System for Disease Control and Prevention in Shanxi . According to our data , 90 . 9% , 91 . 0% , 90 . 3% and 97 . 8% HFMD cases were children aged 0–5 years in Datong , Taiyuan , Changzhi and Yuncheng , respectively . Therefore , we focused analysis on the incidence of HFMD among children aged 0–5 years in this study . The analysis includes descriptive , correlation and time series regression analyses . The meteorological variables data were calculated for intervals of 7 consecutive days , and transformed into a time series format . Descriptive analysis was performed by describing the distribution of climate variables and HFMD cases . Spearman rank correlation and partial correlation analysis were used to examine the association between each meteorological variable and the incidence of HFMD . In addition , given the potential lagged effect of the meteorological variables on disease transmission , cross-correlation analysis was also performed with relevant time lag values . Time series analysis was used to assess the effect of climatic variables on HFMD incidence . The plot of the observed HFMD incidence showed most of the cases occurred from May to July in the 4 cities ( Fig . 2A-2D ) . Furthermore , the plots of autocorrelation function ( ACF ) and partial auto correlation function ( PACF ) of HFMD cases ( Fig . 2E-2H ) showed the time series was non-stationary . With the temporal dependence of HFMD incidence , the need to use a SARIMA model was evident . The Seasonal Autoregressive Integrated Moving Average ( SARIMA ) model ( Box and Jenkins method ) has been recently applied in epidemiological studies [24 , 30–33] , and was used to describe current ( and future ) incidence of HFMD in terms of their past values in this study . SARIMA models extend basic ARIMA models and allow for the incorporation of seasonal patterns . A SARIMA model , which includes seasonal and non-seasonal components , is typically represented by ( p , d , q ) ( P , D , Q ) s , where p represents the order of autoregression ( AR ) , d is the order of differencing , and q is the order of the moving average ( MA ) . P , D , and Q are their seasonal counterparts , and s is the seasonal lag [34] . The long-term trend and seasonal components of each time series can be removed using SARIMA models [24] . In addition , we considered the weather during holidays may also affect the occurrence of HFMD . The development of a SARIMA model is a four-step process . Therefore , for the HFMD time series analysis , it was firstly necessary to stabilize the variance of the series by square root transformation , and seasonal and regular differencing was also applied . Secondly , in order to identify the order of MA and AR parameters , the structure of temporal dependence of stationary time series was assessed respectively , by the analysis of autocorrelation ( ACF ) and partial autocorrelation ( PACF ) functions . From the correlograms of the series , the p value may equal 0 , 1 or 2 for autoregressive parameters and q value may equal 1 , 2 or 3 for moving average parameters ( Fig . 3A-3D ) . Thirdly , parameters of the model were estimated by using the maximum likelihood method . The goodness-of-fit of the models was determined for the most appropriate model ( the lowest normalized Bayesian Information Criteria ( BIC ) and the highest stationary R square ( R2 ) ) , using the Ljung-Box test that measures both ACF and PACF of the residuals , which must be equivalent to white noise . The significance of the parameters should be statistically different from zero . Finally , the predictions were performed by using the best fitting model . The predictive validity of the models was evaluated by calculating the root mean square error ( RMSE ) , which measures the amount by which the fitted values differ from the observed values . The smaller the RMSE , the better the model is for forecasting . Therefore , the SARIMA model was developed and verified by dividing the data file into two date sets: the data from the 1st calendar week of 2009 to the 52nd calendar week of 2012 were used to construct a model; and those from the 1st calendar week to the 52nd calendar week of 2013 were used to validate it . For statistical analysis SPSS version 19 . 0 and Stata version 12 . 0 were used . The sensitivity analysis was conducted based on daily unit to check whether the number of weekly HFMD cases could affect the result estimates . Meanwhile , we controlled for day of the week and public holidays using categorical indicator variables . In addition , we graphically examined the exposure-response curves derived using a smoothing function [28 , 35–37] , and natural cubic splines [35] to control long-term trend and seasonality with 6 df per year for time [37] , which was done using the distributed lag non-linear models ( dlnm ) package in the software R .
Meteorological variables and number of HFMD cases show differences from the northern city to the southern city ( Table 1 ) . The northern city ( Datong ) has a lower temperature and relative humidity than the central city ( Taiyuan ) and southern cities ( Changzhi and Yuncheng ) ; while the southern cities have less sunshine than the central city and northern city . During the study period , the number of HFMD cases in Taiyuan and Changzhi were more than that in Datong and Yuncheng . In the 4 study cities , precipitation , temperature , relative humidity and hours of sunshine were positively correlated with incidence of HFMD ( p < 0 . 05 ) . Different meteorological variables may also be correlated with each other . For example , average temperature was positively correlated with maximum temperature in the 4 cities from north to south ( rs = 0 . 994 , 0 . 990 , 0 . 990 , 0 . 981; p < 0 . 001 , respectively ) , and also correlated with minimum temperature ( rs = 0 . 993 , 0 . 988 , 0 . 987 , 0 . 989; p < 0 . 001 , respectively ) . Accounting for these correlations , the association between meteorological variables and the number of HFMD cases were then analyzed using partial correlations . Results showed the associations of increased number of HFMD cases with increasing atmospheric temperature in the 4 cities ( p < 0 . 05 ) . In addition , the results showed statistically significant but weaker correlation for the association between relative humidity and the incidence of HFMD in Taiyuan , as well as the weaker correlation between precipitation and HFMD in Changzhi ( Table 2 ) . In order to estimate the values of parameters in fitted models , these models were diagnosed by analyzing the data with several SARIMA models without the weather variables , and the models in which the residual was not likely to be white noise were excluded . Therefore , the univariate SARIMA ( 0 , 1 , 1 ) ( 2 , 0 , 1 ) 52 model for Datong; SARIMA ( 2 , 1 , 3 ) ( 1 , 1 , 1 ) 52 model for Taiyuan; SARIMA ( 0 , 1 , 1 ) ( 0 , 1 , 1 ) 52 model for Changzhi; and SARIMA ( 0 , 1 , 1 ) ( 1 , 1 , 2 ) 52 model for Yuncheng had both the lowest Bayesian information criterion ( BIC ) and the highest R2 values and were the best to fit the HFMD cases , respectively ( Table 3 ) . The Ljung-Box test confirmed that the residuals of the time series were not statistically dependent ( p > 0 . 05 ) and the residuals on ACF and PACF plots showed the absence of persistent temporal correlation ( Table 3 ) ( Fig . 4 ) . The selected SARIMA model fitted the observed data from 2009 to 2012 . Then , the model was used to project the number of HFMD cases between January to December 2013 , and was validated by the actual observations . The validation analysis suggested that the model had reasonable accuracy over the predictive period in the 4 cities ( root-mean-square error ( RMSE ) = 1 . 763 , 4 . 505 , 4 . 907 and 2 . 817 , respectively ) ( Table 3 ) . The cross-correlation analyses showed the lag effects of the meteorological variables on the number of HFMD cases were different in the 4 cities . In Datong , HFMD was significantly positively associated with average temperature at lag 0 ( coefficients = 0 . 540 , p < 0 . 05 ) , mean maximum temperature at lag 0 ( coefficients = 0 . 520 , p < 0 . 05 ) , mean minimum temperature at lag 0 ( coefficients = 0 . 542 , p < 0 . 05 ) . In Taiyuan , the cases were significantly positively associated with average temperature at lag 1 week ( coefficients = 0 . 615 , p < 0 . 05 ) , maximum temperature at lag 1 week ( coefficients = 0 . 612 , p < 0 . 05 ) , minimum temperature at lag 1 week ( coefficients = 0 . 606 , p < 0 . 05 ) , relative humidity at lag 3 weeks ( coefficients = 0 . 224 , p < 0 . 05 ) . In Changzhi , the disease was significantly positively associated with average temperature , maximum and minimum temperature at lag 1 week ( coefficients = 0 . 479 , 0 . 472 and 0 . 465 , p < 0 . 05 ) , respectively . In Yuncheng , HFMD was significantly positively associated with average temperature at lag 1 week ( coefficients = 0 . 351 , p < 0 . 05 ) , maximum temperature at lag 2 weeks ( coefficients = 0 . 347 , p < 0 . 05 ) and minimum temperature at lag 1 week ( coefficients = 0 . 352 , p < 0 . 05 ) . To reduce potential multicollinearity , weekly average temperature , weekly mean maximum temperature and weekly minimum temperature were put into separate regression models ( Models 1 , 2 and 3 ) . In Model 1 , 2 and 3 , temperature ( average , maximum , and minimum , respectively ) was included , along with other meteorological variables . The results indicated that average temperature , and maximum and minimum temperatures with different lag times were significant in the SARIMA Model 1 , Model 2 and Model 3 , respectively ( Table 4 ) . Overall , SARIMA models with temperature were a better fit and validity than the models without the variable ( Stationary R-squared ( Stationary R2 ) increased , while the BIC decreased ) ( Table 3 and Table 4 ) . Other meteorological variables were not significantly included in the models , indicating their contribution was not statistically significant in this study . The models suggest that in Datong , a 1°C rise in weekly average temperature , weekly mean maximum temperature and weekly mean minimum temperature may be related to an increase in the weekly number of cases of HFMD of 0 . 8% ( 95%CI: 0 . 3%-1 . 2% ) , 0 . 6% ( 95%CI: 0 . 1%-1 . 0% ) and 0 . 7% ( 95%CI: 0 . 2%-1 . 3% ) respectively . In Taiyuan , a 1°C rise in weekly average temperature , weekly mean maximum temperature and weekly mean minimum temperature may be related to an increase in the weekly number of cases of HFMD of 1 . 4% ( 95%CI: 0 . 5%-2 . 7% ) , 1 . 0% ( 95%CI: 0 . 3%-1 . 8% ) and 1 . 1% ( 95%CI: 0 . 2%-2 . 1% ) respectively . In Changzhi , a 1°C rise in weekly average temperature , weekly mean maximum temperature and weekly mean minimum temperature may be related to an increase in the weekly number of cases of HFMD of 1 . 1% ( 95%CI: 0 . 1%-2 . 1% ) , 1 . 6% ( 95%CI: 0 . 4%-2 . 7% ) and 1 . 5% ( 95%CI: 0 . 3%-2 . 7% ) respectively . Finally , in Yuncheng a 1°C rise in weekly average temperature , weekly mean maximum temperature and weekly mean minimum temperature may be associated with an increase in the weekly number of cases of HFMD of 2 . 1% ( 95%CI: 0 . 4%-4 . 7% ) , 1 . 4% ( 95%CI: 0 . 3%-8 . 4% ) and 1 . 9% ( 95%CI: 0 . 1%-5 . 8% ) respectively ( Table 4 ) . Table 4 indicates that the incidence of the disease may rise with the increase of temperature , but only within a certain range of temperatures . The selected SARIMA model was used to project the number of HFMD cases in each city for the 52 weeks between January and December 2013 . The validation for January to December 2013 data showed a good fit between observed and predicted data ( Fig . 5 ) . A comparison between the models for the 4 cities , reveals that although a 1°C rise in temperature may cause a similar relative increase in the number of cases , the lag times for the effects of temperatures were shorter in Datong ( at lag 0 ) than those in Taiyuan , Changzhi and Yuncheng ( at lag 1 week ) . The lag times for the effects of maximum temperature were longer in Yuncheng ( at lag 2 weeks ) than those in Changzhi , Taiyuan and Datong ( Table 4 ) . In the sensitivity analysis , the results using daily data indicated that a 1°C rise in daily average temperature may be related to an increase in the daily number of cases of HFMD of 1 . 2% ( 95%CI: 0 . 1%-2 . 3% ) at lag 1 day , 1 . 6% ( 95%CI: 1 . 0%-2 . 2% ) at lag 8 days , 1 . 5% ( 95%CI: 0 . 5%-2 . 5% ) at lag 6 days , and 2 . 4% ( 95%CI: 0 . 1%-4 . 7% ) at lag 8 days in Datong , Taiyuan , Changzhi and Yuncheng , respectively . Although the analysis of daily data is likely to yield a more precise estimate compared to weekly data , the results were similar . Fig . 6 shows non-linear dose-response relationships for temperature with HFMD occurrence in the 4 cities , and an increase in HFMD occurrence within a short interval .
Although there are some time series studies in China on HFMD , these studies scarcely considered the potential impact of socioeconomic status . The findings in this study indicate that the occurrences of HFMD were positively associated with temperature in 4 major cities which have similar socioeconomic status and demographic characteristics . In addition , different lag effects of temperature were observed in selected regions from north to south . The results will be useful to assist public health responses in the different regions and informing local community and health authorities to better predict disease outbreaks . | Understanding of the impact of weather variables on HFMD transmission remains limited due to various local climatic conditions , socioeconomic status and demographic characteristics in different regions . This study provides quantitative evidence that the incidence of HFMD cases was significantly associated with temperature in Shanxi Province , North China . The delayed effects of weather variables on HFMD dictate different public health responses in 4 major cities in Shanxi Province . The results may provide a direction for local community and health authorities to perform public health actions , and the SARIMA models are helpful in the prediction of epidemics , determination of high-risk areas and susceptible populations , allocation of health resources , and the formulation of relevant prevention strategies . In order to reduce future risks of climatic variations on HFMD epidemics , similar studies in other geographical areas are needed , together with a longer study period to enable trend analysis which takes into consideration local weather conditions and demographic characteristics . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | The Effect of Meteorological Variables on the Transmission of Hand, Foot and Mouth Disease in Four Major Cities of Shanxi Province, China: A Time Series Data Analysis (2009-2013) |
Chromosomal gains and losses comprise an important type of genetic change in tumors , and can now be assayed using microarray hybridization-based experiments . Most current statistical models for DNA copy number estimate total copy number , which do not distinguish between the underlying quantities of the two inherited chromosomes . This latter information , sometimes called parent specific copy number , is important for identifying allele-specific amplifications and deletions , for quantifying normal cell contamination , and for giving a more complete molecular portrait of the tumor . We propose a stochastic segmentation model for parent-specific DNA copy number in tumor samples , and give an estimation procedure that is computationally efficient and can be applied to data from the current high density genotyping platforms . The proposed method does not require matched normal samples , and can estimate the unknown genotypes simultaneously with the parent specific copy number . The new method is used to analyze 223 glioblastoma samples from the Cancer Genome Atlas ( TCGA ) project , giving a more comprehensive summary of the copy number events in these samples . Detailed case studies on these samples reveal the additional insights that can be gained from an allele-specific copy number analysis , such as the quantification of fractional gains and losses , the identification of copy neutral loss of heterozygosity , and the characterization of regions of simultaneous changes of both inherited chromosomes .
DNA copy number aberration ( CNA ) , defined as gains or losses of specific chromosomal segments , are an important type of genetic change in tumors . Various microarray based experimental platforms [1]–[7] have made possible the fine scale measurement of CNAs . Whereas the earlier platforms such as comparative genome hybridization arrays were designed to measure the total copy number of both inherited chromosomes , other platforms such as high density genotyping microarrays [6]–[8] can measure allele specific DNA quantity . For alleles that represent known variants of genes , it would be of biological interest to know which allele has undergone copy number change [9] . Also , some genetic mechanisms , such as gene conversion , mitotic recombination , and uniparental disomy , cause loss of heterozygosity ( LOH ) without change in total DNA copy number , and thus can not be detected through conventional analysis methods relying only on total copy number . Even in the case where the total DNA copy number changes , it would be informative to know whether one or both of the inherited parental chromosomes are involved . Thus , to construct a more detailed molecular portrait of tumors , we need to distinguish between the underlying quantities of the two inherited chromosomes , which we call the parent specific copy numbers . This paper addresses the problem of parent specific copy number estimation using allele-specific raw copy number data from high-density genotyping arrays . We will describe the data in more detail in the next section . Here , we clarify the differences between total copy number analysis and parent specific copy number analysis , and review the background of the computational treatment of this problem . The genome of each somatic human cell normally contains two copies of each of the 22 autosomes , one inherited from each biological parent . At any genome location , one or both of these two chromosomes may gain or lose copies , thus creating a change in total copy number at that location . Microarray experiments for measuring total copy number produce a sequence of continuous valued measurements mapping to ordered locations along the chromosomes . Computational methods can be applied to segment this noisy sequence of measurements into regions of homogeneous copy number [10]–[21] , see Lai and Park [22] and Willenbrock and Fridlyand [23] for a review . Since chromosomes are gained and lost in contiguous segments , the true total copy number should be piecewise continuous . This is why change-point models and hidden Markov models have been very useful for total copy number estimation . Total copy number estimates do not reveal which ( or both ) of the two inherited chromosomes have been gained or lost , and if a locus is polymorphic , which ( or both ) of the alleles have been affected . This information is now available in data produced by high density genotyping platforms , which give , at selected single nucleotide polymorphisms ( SNPs ) , a bivariate measurement quantifying the two alleles which we arbitrarily label and , as shown in the left panel of Figure 1 . Some platforms output the total raw copy number ( ) , which is the sum of and , and the B-allele frequency ( BAF ) , which is the percentage of allele raw copy number among the total allele raw copy number , i . e . , . The logR quantifies the total copy number , while the BAF quantifies the imbalance between the two alleles . The right panel of Figure 1 shows , the sum of and allele intensities , and BAF . Unlike the total copy number , the allele-specific measurements are mixtures that depend on the unknown genotype at each location . For this reason , conventional change-point models can not be applied to allele specific copy number estimation . This problem can be formulated statistically as follows: The observed and intensities form a bivariate sequence whose underlying distribution undergoes abrupt changes . The distributions at each location are mixtures . Both the change-points , the mixture components , and the cluster memberships at each data point are unknown and must be estimated from the data . There have been much effort extending existing genotyping and total copy number segmentation procedures to analyze allele-specific data . At the probe level , CNAT [24] , CN5 [24] , CRMA [25] , dChipSNP [26] , [27] , PLASQ [28] , and PICR [29] can be applied to Affymetrix data to produce allele-specific probe-set summaries at each SNP location . However , just as in the estimation of total copy number , the allele-specific intensities for adjacent SNPs should be smoothed to infer the underlying parent-specific copy numbers . LaFramboise et al . [28] first segmented the total copy number using Circular Binary Segmentation [30] , and then estimated the parent-specific copy numbers for each segment . This early approach misses copy neutral loss-of -heterozygosity ( LOH ) events , defined as the simultaneous gain of one chromosome and balanced loss of the other chromosome resulting in loss of heterozygosity but no change in total copy number . Many other existing approaches rely on discrete-state hidden Markov models [27] , [31]–[34] , which are hidden Markov models assuming a pre-specified finite set of underlying states . For example , PennCNV [32] and QuantiSNP [33] assume that the underlying copy numbers belong to the integer classes , and that the allele-specific copy numbers can be described by “generalized genotypes” AA , AB , BB , A- , B- , AAB , ABB , etc . While these types of models are very useful for detecting germline copy number variants in normal tissue , they do not generalize well to genetically heterogeneous samples . This is because by requiring a fixed set of pre-defined discrete states , they do not account for the heterogeneity of cells within the sample , which produces data with apparently fractional copy number changes rather than the idealized unit-copy changes . This is especially problematic for tumor samples , which are usually heterogeneous mixtures of cells with different genetic profiles . Through titration studies , Staaf et al . [35] showed that methods relying on idealized genotype states lose sensitivity when tumors are diluted with normal cells . The fractional changes in tumors inspired recent approaches [35] , that segment both the logR and BAF simultaneously . Since BAF is a mixture of homozygous and heterozygous SNPs , it cannot be processed using existing segmentation procedures . Current methods solve this problem through a pre-processing step that gets rid of the homozygous SNPs . However , identifying the “homozygous SNPs” is nontrivial when the regions of CNA are unknown , and a segmentation procedure that simultaneously genotype each SNP while inferring the underlying parental copy numbers is desirable , unless a matched normal is available . In light of these recent developments , we need a systematic stochastic model for parent specific copy number which can accommodate fractional copy number changes . We propose a general two-chromosome hidden Markov model for this problem . The hidden states of the model represent the copy numbers of each of the two inherited chromosomes , and take value in the continuous space of real numbers . Thus , unlike discrete state space HMMs , this model is not limited to idealized unit-copy changes . Computationally efficient fitting algorithms are given that scale well to data obtained from the current high density genotyping arrays . The estimation procedure based on the two chromosome model , which we call Parent-Specific-Copy-Number ( PSCN ) , extends the framework developed in Lai et al . [37] for total copy number analysis . After segmenting the genome into regions of constant parent-specific copy number , we identify , for each region , whether both or only one of the parental chromosomes have changed copies . We also determine , in regions containing simultaneous gain of one chromosome and loss of the other , whether the changes are balanced . Thus , we classify the regions into six different types of aberrations depending on the status of the two parental chromosomes: gain of both chromosomes ( gain/gain ) , gain of only one chromosome ( gain/normal ) , gain of one chromosome and balanced loss of the other chromosome ( balanced gain/loss ) , gain of one chromosome and unbalanced loss of the other chromosome ( unbalanced gain/loss ) , loss of only one chromosome ( normal/loss ) and loss of both chromosomes ( loss/loss ) . To our knowledge , this is the most detailed classification available among methods for allele-specific analysis . The PSCN method outputs the copy number for both chromosomes in each segment . We evaluate the accuracy of the proposed procedure on a series of simulated tumor titration data provided by Staaf et al . [35] , as well as a new set of simulation data containing a larger variety of chromosomal aberrations . We then apply the new approach to 223 glioblastoma samples from the Cancer Genome Atlas project [38] , and illustrate through case studies some of the insights gained from an analysis of allele-specific data .
Let be the allele-specific signals for alleles A and B at SNPs ordered by their locations in a reference genome . The way of obtaining depends on the experimental platform ( see “Data Transformation” in Methods ) . Our goal is to infer the quantities of the parent specific copy numbers , which we denote by . By parent-specific , we distinguish between the chromosomes inherited from the two parents , which we treat as exchangeable and do not label as maternal or paternal . Let be the configuration at SNP specifying the alleles carried by the inherited chromosomes . Let be the true copy numbers of alleles and at SNP . The relationship between , , and is shown in Table 1 . Note that when a somatic event causes a change in copy number of one or both parental chromosomes at SNP , the allele-specific copy numbers change , but remains fixed . For example , if the inherited genotype is , and if is amplified two-fold , then the true copy number of allele would also be amplified two-fold , but would still be . The observed allele specific signals are assumed to be equal to the true allele specific quantities plus an independent measurement error , ( 1 ) where and are state specific error covariance matrices . The model that relates to , and is illustrated in Figure 2 . To model the gains and losses of the two inherited chromosomes , we assume that is a Markov jump process with state space . Conceptually , each time jumps , it can choose between two states: The normal state ( one copy each of maternal and paternal chromosome ) , where must assume a known baseline value , or the variant state , where picks a new random value from the bivariate Gaussian . The prior mean and prior covariance , along with the other hyperparameters of the prior , will be estimated by maximum likelihood . To allow the possibility of the copy number changing from a variant state to a different variant state , for example , to , we technically need two identically distributed variant states in our formulation of the Markov chain . Hence we let the states be . Then , the dynamics of the Markov model can be described by the transition matrix ( 2 ) The matrix specifies that if is in the normal state at SNP , then at SNP , stays in the normal state with probability , or jumps to a variant state with probability . If is in a variant state , then at SNP , it would stay at the same variant state with probability , or jump to a different variant state with probability , or jump back to the normal state with probability . One can verify that this formulation of the Markov chain , with one baseline state and two variant states , allows for a model with a baseline state and generic “variant” states as desired . This model extends the one used for the analysis of total copy number in Lai et al . [37] . This Markov chain has the stationary distribution . The three-state Markov chain with transition probability matrix and initialized at the stationary distribution is reversible , which provides substantial simplification for the estimation of . Practically , the reversibility of the Markov model implies that we would obtain the same segmentation going from right to left as we do going from left to right . Biologically , this seems logical , as there is no known directionality of copy number aberration events . We assume that the inherited allele configurations are independent multinomial with prior parameterswhich can be obtained from the genotyping data of a set of normal control samples . Note that and cannot be distinguished in normal samples , so we can set and to one-half of the proportion of heterozygotes for SNP . When these figures are not available , we have found that a uniform prior usually works reasonably well . This is because the main purpose of the model is to estimate the parent-specific copy numbers , with as surrogate information . With the large number of data points obtained from the high density arrays , the posterior for the parent-specific copy numbers is usually quite insensitive to the prior on . Note that for platforms , such as the Affymetrix 6 . 0 array , have non-polymorphic copy number markers rather than SNP markers . For those markers , the prior for can be set to . In this way , the posterior will always remain at and only the total copy number information at these markers would contribute to the overall segmentation . Note that this model contains many assumptions , including Gaussianity of the allele specific intensities and Markovicity of the underlying copy number states . These assumptions allow fast and explicit analytic formulas to be derived , thus avoiding the need for Monte Carlo based estimates . For most platforms , the allele-specific intensities deviate from Gaussianity , despite careful normalization . Also , there has never been proof that chromosomal breakages are Markovian . These assumptions are made for modeling convenience , just as in the total-copy number estimation problem [11] , [16] , [30] , [37] . It is reassuring that the estimation method is robust to deviations from both the Gaussian and Markov assumptions , as we show using the titration data from Staaf et al . [35] and through our own spike-in studies . Our primary objective is to estimate the parent specific copy numbers , which depend on the observed signals through the unobserved inherited allele configurations . Let and be the set of all possible realizations for and , respectively . We describe below an iterative algorithm to estimate and . The segmentation divides the genome into regions where the copy numbers of the two inherited chromosomes are constant . It is often useful to know , for each region , whether the copy numbers of one or both parental chromosomes deviate from the normal level . This involves classifying each region into one of the following six types of chromosomal change: gain/gain , gain/normal , balanced gain/loss , unbalanced gain/loss , normal/loss and loss/loss . For each segmented region , we define the major copy number to be the normalized raw copy number of the more abundant chromosome , and the minor copy number to be the normalized raw copy number of the less abundant chromosome . If the two chromosomes have equal copy numbers , then the major and minor chromosome labels are assigned arbitrarily . The major and minor copy numbers are estimated after the hard-segmentation using a mixture model on the heterozygous SNPs in each region ( which can be identified using ) . Then , a -test is used to compare the estimated major and minor copy numbers of each region to the estimated allele copy number of the normal level in the unchanged segments . The Bonferroni correction is used to adjust for multiple testing . The technical details are given in Methods . This procedure allows us to discover and distinguish all of the six types of CNVs . An additional caveat is that when both parental chromosomes carry the same haplotype , a balanced gain/loss would be called if the region were long enough . Without matched data from normal tissue , it is impossible to distinguish with certainty between inherited and somatic LOH . However , we rely on the fact that long regions of LOH are infrequent , and thus the minor allele frequency of SNPs and the linkage disequilibrium between them can be used to conduct a test for the probability that an inherited LOH appears by chance . This haplotype correction only takes care of the unique common haplotypes , i . e . , when a region is dominated by one haplotype . If a haplotype is not common in that region , or if there are several haplotypes in that region , this test loses sensitivity . In this case , paired normal cell information would be useful . More details are given in Methods . Staaf et al . [35] performed a systematic comparison of existing methods for allele-specific copy number estimation . They created a simulated dilution data set based on experimental 550k Illumina data for HapMap sample NA06991 . To the diploid HapMap sample , ten regions of aberrant copy number were added at increasing fractions to mimic a tumor sample that is contaminated with normal cells . Here , normal cell contamination means part normal cells are mixed with part tumor cells . The aberrant regions vary by type and length , and represent regions of hemizygous gains and losses and copy neutral LOH . Since the locations of the true aberrant regions are known , the specificity and sensitivity of the detection methods can be evaluated . We applied PSCN , the R package we developed based on our method , to this dilution data set and compared it with existing approaches in an analysis that parallels the insightful analysis in Staaf et al . [35] . The sensitivity and specificity of results from PSCN at varying contamination ratios is shown in Figures 3 and 4 overlayed onto plots reproduced from Staaf et al . [35] . In order to compare with the sensitivity analysis of other models done in the paper by Staaf et al . [35] , we define a “correct detection” to mean that a true CNA region has been called , but do not require that the type of CNA ( e . g . gain/loss , normal/loss ) has been correctly identified . All the other current procedures only categorize the CNAs into Gain , Loss and LOH , which are the three types of CNAs used in the Dilution data in Staaf et al . [35] . We assess the accuracy of PSCN in a more detailed classification of identified CNAs based on the six types of chromosomal change in a separate data set that contains a wider diversity of chromosomal events ( see next section ) . In the simulated dilution data , the regions vary in length , magnitude , and type of aberration , with some regions harder to detect than the others . There is a separate sensitivity plot for each of the 10 aberrant regions created by [35] . As expected , for all regions , sensitivity is maintained at a high level up to a certain contamination ratio , then drops sharply . Since Staaf et al . and we used very stringent detection thresholds , the specificity is maintained near 1 for all contamination ratios , as shown in Figure 4 . The sensitivity of PSCN is comparable to SOMATICs [36] , but the latter method has much lower specificity , as shown in the analysis of Staaf et al . , see Figure 4 . PSCN achieves good accuracy compared to the other existing methods , especially methods based on discrete-state hidden Markov models for high levels of contamination . The discrepancy between the two specificity plots in Figure 4 are due to the fact that when an aberration is called , it may be labeled as an incorrect type ( for example , a copy neutral LOH may be labeled as single copy gain ) . When the correct calling of aberration type is required , the specificity of PSCN is maintained through a higher level of contamination as compared to existing models . The new model can identify the correct aberration type if the normal cell contamination is below 80% . Above 80% , PSCN gains significantly in sensitivity compared to existing methods but also sacrifices slightly in specificity . The dilution data set from Staaf et al . [35] contains only three types of aberrations: hemizygous loss ( normal/loss ) , single copy gain ( gain/normal ) , and copy neutral LOH ( balanced gain/loss ) . We created a simulated data set containing all six types of aberrations: gain/gain , gain/normal , balanced gain/loss , unbalanced gain/loss , normal/loss and loss/loss . To make the simulation resemble real data , we started with the 550k Illumina data for chromosome 1 of HapMap sample NA06991 . To this normal sequence we imposed six different signal types on six regions . The positions and magnitudes of the added signals are shown in Table 2 . The top panel of Figure 5 ( first row ) shows the and BAF before the signals are imposed . The middle and bottom panels show the and BAF after the signals have been imposed , at 0% and 80% contamination respectively , with true signals indicated by black lines . Signal becomes weaker when normal cell contamination increases , and thus are harder to detect . The estimated parent-specific copy numbers are shown in Figure 6 . We can see from the plots that the estimated parent-specific copy numbers are very close to the true allele copy numbers . Table 3 shows the largest normal cell contamination under which the signals are detectable by PSCN . When normal cell contamination is less than 80% , our model can detect most of the signals with both alleles assigned to the correct type . When the normal cell contamination rises to 90% , our model can still detect three out of the six CNA regions , but assigns the correct type to only one of the two alleles . For example , at a high contamination level of 90% , there is a tendency for a fractional loss of both chromosomes to be mistaken for a fractional loss of only one of the two chromosomes . From this study , we see that the correct type of aberration can be identified robustly for all but the highest levels of normal cell contamination . Using the dilution data set created from HapMap sample NA06991 , we can also assess the accuracy of PSCN in identifying the genotype states . Since the genotypes for the SNPs on this sample are known , we simply compared the estimated with the true values . Table 4 shows the percent of homozygous SNPs that are misclassified as heterozygous , and vice versa . When the SNP is classified as homozygous , the determination between the states AA and BB is trivial , and no errors are made . When normal cell contamination is extremely low , less than 10% , genotyping errors are common in regions of loss of heterozygosity ( either normal/loss or gain/loss ) . This is expected , since in a region with complete LOH and zero contamination , only one of the two parental alleles is left , and thus it would be impossible to distinguish between the homozygous configurations and the heteryzogous configurations . Fortunately , these types of genotyping errors would not affect the accurate estimation of , since the mean levels for the heterozygous and homozygous tracks merge for LOH regions under zero contamination . It is slightly unintuitive that the correct estimation of depends on the fact that there is normal cell contamination ! This is reflected in Table 4 , where accuracy quickly improves as normal cell contamination increases , with a total misclassification rate of at normal cell contamination . A complete analysis of the misclassification rates of are given in the Supporting Information file ( Text S1 ) . We applied PSCN to 223 glioblastoma samples from the TCGA project [38] . These samples were assayed using Illumina HumanHap 550k SNP arrays . Almost all of the 223 samples analyzed contain substantial copy number aberrations . Table 5 shows the distribution of the types of copy number events found in the samples . Of the gain/loss events , which comprise 45 . 4% of all of the events , 22 . 8% are copy neutral LOH and 22 . 5% are unbalanced gain/loss . We see from this table that , among these glioblastoma samples , single chromosome losses or single chromosome gains comprise 49 . 6% of all the events , which means that more than half of the events involve change of both inherited chromosomes . We now zoom in on two example regions to illustrate the additional insights gained from parent-specific copy number analysis . These regions are shown in Figure 7 . The figures in the left panel correspond to the entire chromosome 3 of TCGA glioblastoma sample 02-0332 , while those on the right panel correspond to the first 10000 SNPs on chromosome 2 of TCGA glioblastoma sample 02-0258 . The top two plots in each panel show the and BAF values . The color scheme for these plots show the segmentation obtained using PSCN . We transformed the and BAF values back to the raw copy number values , and fitted two dimensional densities separately to each region in the segmentation . The contours of the two dimensional density estimates , delineating the locations of the clusters , are shown in the third plot from the top in each panel . The color scheme for the contours is the same as the color scheme for the and BAF plots . Finally , the bottom plot of each panel shows the estimated major and minor copy numbers for each region ( we will call this type of plot the mm-plot ) . The color scheme of the mm-plot reflects the gain/loss status of each region , where red represents gain , blue represents loss , and green represents normal . It is usually difficult to discern the relative magnitudes of gains and losses from the and BAF plots , especially when both inherited chromosomes have undergone copy number changes . Such relative changes in parent specific copy numbers can be quantified more easily by examining the contour and -plots .
We have developed a method for simultaneous estimation of parent-specific DNA copy number and inherited genotypes for tumor samples using allele-specific raw copy number data . The model and estimation procedure start with transforming allele-specific data into and intensities , which may vary across experimental platforms . The model assumes that the and allele intensities should be roughly symmetric , roughly variance stabilized and have approximately bivariate Gaussian errors . Indeed , the model is quite robust to the violation of the bivariate Gaussian error assumption . The model gives satisfying results even if this assumption is heavily violated . More details are shown in the the Supporting Information file ( Text S1 ) . We illustrated the method and evaluated its performance on both published and newly generated dilution data sets on the Illumina platform . A rigorous assessment using in silico titration data provided by Staaf et al . [35] shows that PSCN has good accuracy . The proposed method does not require paired normal samples . However , if such samples were available , then they can be used to further improve accuracy and to distinguish between inherited LOH and somatic LOH . In such cases , can simply be set to the genotypes inferred from the normal samples . PSCN is not platform specific , and we have also applied it to data from the Affymetrix Genotyping 6 . 0 array , with an example analysis given in the Supporting Information file ( Text S1 ) . The segmentation accuracy of PSCN seems to be reasonable for Affymetrix data , but can potentially be improved significantly by better probe-level normalization . This is due to the fact that the BAF of Affymetrix data is much noisier than the BAF of Illumina data , which makes the estimation of much more difficult . Bengtsson et al . [39] have shown that much of the variation in the BAF of Affymetrix data are due to probe-specific effects that can be removed if a matched normal sample is available . Another promising method for probe-level normalization of Affymetrix data is the probe raw copy number composite representation ( PICR ) model of Wan et al . [29] , which uses probe sequence information and physico-chemical modeling to estimate binding affinity . However , since the PICR model relies on mismatch probes , it is only applicable to Affymetrix platforms prior to the 6 . 0 array . Thus , better probe-level normalization of Affymetrix 6 . 0 data for unmatched samples is still an important problem for further investigation . An overview of an analysis of the TCGA glioblastoma samples reveal that a substantial fraction of copy number changes are copy-neutral loss of heterozygosity events . These events would not have been found using analyses based only on total copy number . Cases of unbalanced simultaneous changes in the copy numbers of both inherited chromosomes were also found . It would be of interest to quantify the frequency of such changes among different cancer subtypes and in other types of tumors . A final point that we would like to emphasize is the quantification of fractional changes , as exemplified by the two case studies on the TCGA glioblastoma samples . Since this requires teasing apart the quantities of the two inherited chromosomes , it can only be achieved through allele-specific estimates . The fraction of cells that carry each copy number event is important for downstream analyses , such as quantifying normal cell contamination and studying tumor microevolution . The parent-specific copy number estimates obtained from the proposed method provides a starting point for these types of investigations . The R package for PSCN is registered on R-Forge ( http://r-forge . r-project . org/ ) under project name PSCN .
The proposed model is not platform specific , and can theoretically be applied to any type of allele-specific copy number data where the errors on the raw copy number values of the alleles can be normalized to approximately adhere to a bi-variate Gaussian distribution . As we show below , the Gaussian error assumption allows for explicit analytic formulas for the posterior mean of the underlying inherited chromosome copy numbers , thus bypassing the need for computationally intensive Monte Carlo methods . For most platforms , the raw allele-specific raw copy number values must be properly normalized for this error model to be a good approximation . However , as we mentioned in the Discussion section , the model is quite robust to the violation of the Gaussian error assumption . A unified approach that gives satisfying results for data from both Illumina and Affymetrix platforms is as follows . Sincewe haveNote that the “BAF” given by the Illumina platform [6] is not the intuitive quantity ( ) , but the arc-tangent of the ratio of raw copy number versus raw copy number scaled to [0 , 1] . Use to denote the so called BAF given by Illumina , thenFor PSCN we use . We give here exact formulas for the conditional expectation ( 3 ) . Let denote the probability distribution that assigns probability 1 to the value . Denote by , and . A brief outline of the estimation procedure is as follows: First , conditioned on all data to the left of , is distributed as a mixture of Gaussians: ( 5 ) where the formulas for computing the parameters of the mixture , , and are given below . We call ( 5 ) the forward filter . Since by our model is a reversible Markov chain , we can reverse time and obtain a backward filter that is analogous to ( 5 ) : ( 6 ) where the parameters , , and , as for the forward filter , are given in explicitly computable form below . The Bayes theorem can then be used to combine the forward filter ( 5 ) and backward filter ( 6 ) to derive the posterior distribution of given the complete sequence , which is a mixture of normal distributions ( 7 ) whose parameters can be derived from the forward and backward filters as described below . This forward-backward procedure can be reduced to computation time by the BCMIX algorithm [40] . From ( 7 ) , it follows that the conditional expectation in Equation ( 3 ) can be computed as ( 8 ) The variables are assumed to be i . i . d . , withThe inherited allele configurations is assumed to be independent of , so ( 11 ) where is a constant . Each component of the above sum can be maximized separately to give , for each , Let be and intensities of heterozygous SNPs for segments at normal state and be and intensities of heterozygous SNPs for the segment being tested . Then , , follow the model:For the normal state , we can estimated the parameters easily asFor the target segment , , , , can be estimated by EM algorithm: Step 1: Initialize: Step 2: Set Step 3: Set Step 4: Stop if , where is a pre-chosen threshold ( PSCN has default value ) . Otherwise , set , , , , and go back to step 2 . The motivation of the initial and default settings are as follows . For segment with changed states , the goal is to estimate minor and major copy number . It is expected that the minor copy number would be less than or equal to 1 and the major copy number would be larger than or equal to 1 , so the initial values for and are set to 0 . 9 and 1 . 1 respectively . Although it is possible that both chromosomes in a segment are gained or lost , a small discrepancy of the initial values of and will also be a good start . Also , it is expected that the numbers of AB and BA states in a segment is similar , so the initial value of is set to 0 . 5 . The initial values for and can be quite arbitrary , with 1 being a reasonable value to use . is set to be , which is small enough to indicate a convergence of the iterative algorithm . Denote the estimated parameters by , , , , . To test the hypothesis , the standard -statistic isUnder , the distribution of is with degree of freedom , so -value can be calculated and compared with the level of the test . The null hypothesis that needs also be tested , by replacing with in the above equation . | Many genetic diseases are related to copy number aberrations of some regions of the genome . As we know , each chromosome normally has two copies . However , under some circumstances , for some regions , either one or both of the chromosomes change . Genotyping microarray data provides the copy number of the two alleles of polymorphic sites along the chromosomes , which make the inference of the copy number aberrations of the chromosome feasible . One difficulty is that genotyping microarray data cannot provide the haplotype of the two copies of a chromosome . In this paper , we model the copy number along the chromosome as a two-dimensional Markov Chain . Using the observed copy number of both alleles of all the sites , we can determine the parent specific copy number along the chromosome as well as infer the haplotypes of the two copies of the inherited chromosomes in regions where there is allelic imbalance . Simulation results show high sensitivity and specificity of the method . Applying this method to glioblastoma samples from the Cancer Genome Atlas data illustrate the insights gained from allele-specific copy number analysis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"computational",
"biology"
] | 2011 | Estimation of Parent Specific DNA Copy Number in Tumors using High-Density Genotyping Arrays |
Our appreciation for the extent of Epstein Barr virus ( EBV ) transcriptome complexity continues to grow through findings of EBV encoded microRNAs , new long non-coding RNAs as well as the more recent discovery of over a hundred new polyadenylated lytic transcripts . Here we report an additional layer to the EBV transcriptome through the identification of a repertoire of latent and lytic viral circular RNAs . Utilizing RNase R-sequencing with cell models representing latency types I , II , and III , we identified EBV encoded circular RNAs expressed from the latency Cp promoter involving backsplicing from the W1 and W2 exons to the C1 exon , from the EBNA BamHI U fragment exon , and from the latency long non-coding RPMS1 locus . In addition , we identified circular RNAs expressed during reactivation including backsplicing from exon 8 to exon 2 of the LMP2 gene and a highly expressed circular RNA derived from intra-exonic backsplicing within the BHLF1 gene . While expression of most of these circular RNAs was found to depend on the EBV transcriptional program utilized and the transcription levels of the associated loci , expression of LMP2 exon 8 to exon 2 circular RNA was found to be cell model specific . Altogether we identified over 30 unique EBV circRNAs candidates and we validated and determined the structural features , expression profiles and nuclear/cytoplasmic distributions of several predominant and notable viral circRNAs . Further , we show that two of the EBV circular RNAs derived from the RPMS1 locus are detected in EBV positive clinical stomach cancer specimens . This study increases the known EBV latency and lytic transcriptome repertoires to include viral circular RNAs and it provides an essential foundation and resource for investigations into the functions and roles of this new class of EBV transcripts in EBV biology and diseases .
Epstein Barr virus ( EBV ) is a human oncogenic gamma herpesvirus that is carried by greater than 90% of the world’s population . While infection with EBV is generally asymptomatic , the virus persists for the lifetime of the host through an intricate and dynamic interplay with the host immune system , achieved in part through the virus’ utilization of multiple distinct gene expression programs . Initial infection through salivary exchange results in infection of the oral epithelium where the full repertoire of viral “lytic” genes is expressed to facilitate local amplification of virus titers . The virus is then transmitted to circulating naïve B-cells where a “latency type III” viral gene expression program is utilized ( Latent Membrane Proteins ( LMPs ) -1 and -2 , Epstein Barr Nuclear Antigens ( EBNAs ) -1 , -2 , -3A , -3B , -3C , and -LP and the non-coding transcripts , EBER1 , EBER2 , v-sisRNAs , v-snoRNA , RPMS1 , and a set of viral miRNAs encoded within the introns of RPMS1 ) [1–4] . The expression of this full repertoire of latency genes facilitates potent B-cell activation and proliferation , a unique mechanism to expand the infected B-cell population in the host ( i . e . independently of new virus production and de novo infection ) . Once an adaptive immune response is mounted to the antigenic latency proteins , the virus adapts through transitioning to ( and/or selection for ) more restricted latency gene expression programs , type II latency ( EBER1 , EBER2 , RPMS1 , viral miRNAs , EBNA1 and LMP1 and LMP2 ) , type I latency ( EBER1 , EBER2 , RPMS1 , viral miRNAs and EBNA1 ) or type 0 latency ( only non-coding RNAs , EBER1 , EBER2 , RPMS1 , and viral miRNAs ) [1 , 2] where the virus persists at low levels with little detriment to the host . In the context of systemic immune suppression ( e . g . HIV infection or clinically induced ) , variable expression of growth promoting type III latency genes can be better tolerated , which often leads to the development of B-cell lymphomas . In immune-competent individuals the viral utilization of non-coding RNAs in addition to low level type I or type II protein coding latency gene expression , perhaps tolerated through local tumor-immune suppression , provides one or more “hits” towards oncogenic progression with minimal impact from the immune system . The role of EBV latency proteins in contributing to the oncogenic phenotype has been the topic of study for many years . More recent studies have demonstrated the importance of the EBV non-coding RNAs , RPMS1 , EBV miRNAs , and EBER1/2 ( expressed across all EBV infected tumor types ) in contributing to the tumor phenotype through promoting cell cycle progression , inhibiting innate and adaptive immune responses , blocking apoptosis , etc [5–12] . Utilizing non-coding RNAs to modulate host cell signaling pathways is likely an important viral strategy for molding the host cell environment in ways that support of the virus’ needs without eliciting immune clearance . Circular RNAs [13–15] are another conserved class of predominantly non-coding RNAs that have gained increased attention in recent years [16–18] . circRNAs are produced by backsplicing of a 3’ splice donor to an upstream 5’ splice acceptor , generating a covalently closed RNA molecule with increased stability due to their lack of exonuclease susceptible 5’ or 3’ substrates . While our understanding of the roles , functions , and mechanisms of action of circRNAs remains nascent , many studies have pointed to non-coding functions such as acting as microRNA sponges ( e . g . ciRS-7 [19] ) , through cis regulation of transcription [20] , and through modulating RNA splicing [21] . In addition , however , a subset of circRNAs may be translated to form smaller peptide derivatives of the parental genes [22–24] . Given the conserved nature of circRNAs and findings that the majority of circRNAs likely play non-coding regulatory roles , we sought to determine whether EBV utilizes this class of transcripts to help facilitate regulation of host cell signaling without solicitating an adaptive immune response . Here we report a comprehensive analysis of the EBV circRNAome across latency types using a panel of cell lines modeling types I , II and III latency and in B-cell lymphomas and gastric cancers . Further , we report the lytic circRNAome using B-cell receptor cross-linking in two Burkitt’s lymphoma cell lines , Akata and Mutu I . These experiments identified an extensive repertoire of viral circRNAs that may play unique roles in different latency stages of EBV infection and in Burkitt’s and gastric cancers as well as those that may be important in viral lytic replication . Together , this work further expands the repertoire of the viral transcriptome to include the circular RNA class of RNAs and stands to set the stage for the discovery of new mechanisms through which EBV facilitates infection , persistence and oncogenesis .
As an initial exploration into whether EBV expresses circular RNAs , we analyzed our previously reported ribodepletion RNA-seq data from the EBV positive Burkitt’s lymphoma cell line , Akata , induced to undergo viral reactivation through B-cell receptor cross-linking ( GEO series accession number , GSE52490 ) [25] . Using find_circ [26] to identify backsplice junction candidates , over a hundred unique backsplice junction calls were made across the EBV genome using this dataset ( S1 Table and S1 Fig ) . Nevertheless , the bulk of these were low abundance and/or were deemed low-confidence upon realignment of raw RNA-seq data to conjoined sequences spanning a handful of these backsplice junctions and some backsplice junction calls mapped to simple repeat regions raising concerns about misalignment . Further , while for read alignments , the episomal EBV genome was linearized at a region displaying minimal transcription [25 , 27] , we were concerned that some backsplice calls could in fact represent forward splicing across the artificially linearized viral genome end-sequences . To decrease background and enhance the fidelity of circRNA detection , ribodepleted RNAs were treated with RNase R to digest linear RNAs prior to library preparation ( Fig 1A ) . Decreasing the levels of linear RNAs through RNase R digestion also served to increase the circular to linear RNA ratios , thereby increasing the sensitivity of circRNA detection . This was considered important because while linear transcript detection is typically assessed by counting all reads mapping to the entire exonic regions , circRNAs can only be unequivocally assessed by the limited subset of reads spanning the backsplice junction . To achieve a relatively comprehensive assessment of the EBV circRNAome , RNase R-sequencing was performed using cell models representing types I , II , and III latency transcription programs and during reactivation , and was performed using both lymphocyte and gastric cancer cell lines ( Fig 1A ) . For six of these models we also performed ribodepletion-only RNA-seq analysis to help assess the relative performance of RNase R-seq . A global analysis of cellular plus viral reads showed that RNase R treatment resulted in a more than 20-fold higher ratio of backsplice to total splice junction reads , indicative of circular RNA enrichment in RNase R treated samples ( Fig 1B ) . Also significant was the finding that less than 20% of backsplice junction calls from ribodepletion-seq data were enriched in RNase R treated samples by more than 4-fold ( Fig 1C ) . This suggested that a substantial proportion of backsplice junction calls made using ribodepletion-seq data could be false positives . Consistently , of the 148 viral backsplice calls made from ribodepletion-seq data in reactivated Akata cells , only 17 were represented in the RNase R-seq data and 14 of these were enriched by more than 4-fold ( S1 Table and S1 Fig ) . An additional eight unique EBV backsplice junctions with more than 5 reads were detected in the RNase R-seq data from reactivated Akata cells ( S1 Table and S1 Fig ) indicating higher sensitivity with RNase R-seq . Based on these analyses , we conclude that RNase R treatment prior to sequencing provides substantially improved specificity and sensitivity of circRNA detection over ribodepletion alone . Backsplice analysis across all RNase R-seq data sets shown in Fig 1A identified 35 unique EBV backsplice junctions supported by more than 5 reads in at least one sample . For each of these backsplice candidates , we visually assessed junction coverage following re-alignment of each dataset to conjoined sequences spanning each respective backsplice junction ( S1 File ) . Two of these , RPMS1 exon 7 to exon 5_3 and RPMS1 intron 6 ( see S1 File ) likely met the minimum of 6 junction spanning read criteria due to PCR duplicates and were discarded from further analysis . Following this quality check filtering , 33 EBV backsplice junction candidates remained ( Fig 2A ) . While some of the EBV backsplice junctions showed relatively low coverage , others displayed reasonably robust representation ( Fig 2A and S2 Table ) . To help gauge the significance of their expression levels , we plotted viral backsplice junction read counts across each dataset in the context of detected cellular backsplice junction read counts ( Fig 2B ) and we assessed their ranking among all cellular and viral backsplice sites detected ( S3 Table ) . circEBNA_U and circRPMS1_E4_E3a , which were detected in both latency and reactivation conditions ranked 234th and 34th respectively in reactivated Akata cells ( S3 Table ) . The less robustly expressed viral circRNAs , circEBNA_W1_C1 and circEBNA_W2_C1 were ranked 838th and 1679th in JY cells . Notably , although these backsplices were not detected in Mutu III , Jijoye , or reactivated Akata cells through our initial find_circ analysis , they were nevertheless detected in these cell lines following re-alignment to conjoined splice junctions ( Fig 2A ) and by RT-PCR ( see below ) . Lastly , the lytic circRNA , circLMP2_E8_E2 , was ranked 36th in reactivated Akata cells and the lytic circBHLF1 was found to be the 2nd most abundant circRNA in the nuclear/cytoplasmic fractionated reactivated Akata cells ( S3 Table ) . This analysis demonstrated that many of the EBV circRNAs are expressed at levels that are comparable to cellular circRNAs ( Fig 2B ) , raising the contention that they may have a comparable likelihood of functional significance . In type III latency , expression of the EBV nuclear antigens , EBNA-LP , EBNA1 , EBNA2 , EBNA3A , EBNA3B and EBNA3C is derived through alternative splicing of transcripts originating from either the Bam HI C promoter ( Cp; exon C1 ) , or the Bam HI W promoter ( Wp; exon W0 ) ( Fig 3 ) . Each of these transcripts contain multiple pairs of 5’ W1 and W2 exons derived from the Bam HI W repeats that are spliced to downstream exons containing different EBNA reading frames . Wp is primarily utilized during the initial stages of B-cell infection after which the virus switches to Cp utilization . Accordingly , most established type III latency cell lines show exclusive Cp usage ( for example , Mutu III , Jijoye , and JY ( Fig 3 ) ) although some exceptions exist including the IB4 cell line ( Fig 3 ) . In type I and II latency where the only EBNA expressed is EBNA1 , the C or W promoters are epigenetically silenced and EBNA1 is expressed from a third promoter , Qp , located downstream from the Bam HI W repeats ( Fig 3 ) . Because of limitations associated with alignment to repeat sequences , we devised a custom data processing method for analyzing the EBNA latency locus to more accurately assess coverage and splicing across the BamHI W repeat region . Specifically , exon coverage and backsplicing read counts were obtained from alignments to a genome containing the EBNA latency locus with only a single Bam HI W repeat . For display , coverage and backsplice read counts associated with the W1 and W2 exons were divided by the appropriate number of exons/junctions to distribute these counts equivalently across 7 W repeats . A similar approach was used for forward splicing analyses except that two Bam HI W repeats were used for alignments to allow capturing of W2 to W1 splicing events . As shown in Fig 3 , backsplice reads spanning from the W1 and W2 exons to a novel splice acceptor in the C1 exon ( Fig 4A ) were detected in the Cp utilizing latency III cell lines , Mutu III , Jijoye , and JY but not in the Wp utilizing type III latency cell line IB4 , the latency type I cell lines Akata , Mutu I and Sav I , or the latency type II cell lines , SNU719 and YCCEL1 ( note; in Fig 3 , RNase R-seq backsplice read counts are plotted with coverage and forward splice read count data from PolyA-seq data to provide a linear transcript expression context ) . W1 to C1 backsplicing was also observed in reactivated Akata cells , consistent with previous findings that Cp is induced during reactivation in Akata cells [25 , 27] . Although not detected in our initial analysis , EBNA U to W1 exon backsplice reads were also observed in this analysis due to the informatics improvements made to accommodate BamHI W repeat alignment issues ( Fig 3 ) . Using divergent primers in the W1 and C1 exons , both the W1-to-C1 backsplice junction and to a lesser extent , the W1-to-W2-to-C1 forward-and-backslice junctions were captured by RT-PCR ( verified by sequencing ) ( Fig 4B ) . Both transcripts were resistant to RNase R digestion demonstrating the closed circular nature of both the W1-to-C1 and the W2-to-C1 containing RNAs ( Fig 4B ) . Using divergent W2 and C1 primers , only a single band was identified that corresponded to the W2-to-C1 backsplice ( verified by sequencing ) and the associated RNA was resistant to RNase R digestion ( Fig 4B ) . This data validates the expression of both circEBNA_W1_C1 and circEBNA_W2_C1 in type III latency and demonstrates the closed circular nature of both of these transcripts . To assess subcellular localization , W1-to-C1 and W2-to-C1 divergent primer pairs were used to assess nuclear and cytoplasmic fractionated RNAs from Mutu III , Jijoye , and JY cells . This analysis indicated that both circEBNA_W1_C1 and circEBNA_W2_C1 are substantially enriched in the nuclear fractions ( Fig 4C ) ( note: with the predominance of W1-C1 amplification in W1-to-C1 PCR reactions , the W1-to-C1 PCR products likely crossed the amplification threshold cutoff first; Ct values thereby most likely represent W1-to-C1 backsplicing ) . This nuclear localization is consistent with findings that circEBNA_W1_C1 and circEBNA_W2_C1 have little coding capacity . Strikingly , the novel C1 backsplice acceptor for these circRNAs is located only 8 nucleotides downstream from the Cp transcription initiation site ( Fig 4A ) , leaving minimal upstream RNA sequences to bind splicing factors . Backsplicing to this splice acceptor could occur from low level upstream initiated transcripts or possibly through some novel mechanism , such as through interactions between splice donor RNA/protein complexes with promoter engaged ( or proximal ) transcription machinery or DNA . The EBNA U exon is contained within EBNA transcripts driven by all three latency EBNA promoters , Cp , Wp , and Qp . The EBNA U backsplice is detected in types I and III latency ( Fig 3 and S2 Fig ) with the number of backsplice reads roughly reflecting the level of U exon coverage ( Fig 3 ) . No backsplices were detected in the two stomach cancer cell lines , SNU719 and YCCEL1 although this may be due to the very low level of U exon coverage observed in this setting . A fourth promoter located just upstream from Qp , referred to as Fp , gives robust Q-to-U-to-downstream EBNA ORF expression/splicing during reactivation . Accordingly , U exon expression is high during reactivation and consistently , a high level of exon U backsplicing is observed ( Figs 2A and 3 ) . The latency RPMS1 locus is a complex transcriptional unit that gives rise to a set of alternatively spliced long non-coding RNAs and nearly 40 miRNAs that are derived from RPMS1 intronic sequences ( Fig 5 ) [28–30] . Find_circ detected 16 unique backsplice junctions across the different tested conditions ( Fig 2B ) although the many of these were found to be of low abundance . During reactivation , a handful of RPMS1 backsplice junctions were found to be well represented including backsplices from a novel splice donor in exon 7 to exons 5 , 3 , and 2 and backsplices from exon 4 to exons 3a and 2 ( Figs 2 and 5 ) . Of these , circRPMS1_E4_E3a and circRPMS1_E4_E2 were found to be expressed across all latency types ( note: both IB4 and JY contain the B95-8 deletion of exons 1b through 4 and accordingly , no circRPMS1_E4_E3a or circRPMS1_E4_E2 backsplice junctions were detected in these cell lines ) . Using a leftward primer in exon 3a and a rightward primer in exon 4 , we detected both the exon 4 to exon 3a backsplice junction ( validated by sequencing ) and the exon 4 to 2 backsplice junction ( exons 4-to-2-to-3a , validated by sequencing ) ( Fig 6A ) . Both backsplice junctions were detected across latency types although there was no detection in IB4 and JY due to the deletion of these sequences or in Mutu III cells which displays low expression of this locus . Further , both fragments were detected in RNase R treated RNAs , validating their closed circular nature ( Fig 6A ) . To determine whether the circRPMS1_E4_E3a and circRPMS1_E4_E2 circular RNAs could be detected in the patient tumor setting , RT-PCR was performed using RNAs from two EBV positive stomach cancer patient specimens that we used in previous studies [31 , 32] . As shown in Fig 6B , both circRPMS1_E4_E3a and circRPMS1_E4_E2 were detected in each of these tumor samples demonstrating their expression in the natural patient tumor setting . Analysis of nuclear/cytoplasmic distribution using exon 4 and exon 3a divergent primers in SNU719 and YCCEL1 cells where the circRPMS1_E4_E3a PCR signal predominates over the circRPMS1_E4_E2 signal ( Fig 6A ) revealed potential low level cytoplasmic localization but with predominant nuclear localization ( Fig 6C ) . With the upstream splice acceptor for circRPMS1_E4_E3a flanking a microRNA cluster intron , circRPMS1_E4_E3a backsplicing could influence the regulation of microRNA processing , although such a scenario doesn’t preclude other possible nuclear functions of mature circRPMS1_E4_E3a RNA . While the LMP2 gene is typically expressed in type III latency cells but not in latency type I cells , reactivation in type I latency cells broadly activates the expression of latency genes , including LMP2A [33] . We have previously demonstrated that in contrast to the normal sequential forward splicing typically observed for LMP2A in type III latency , the LMP2 locus exhibits extensive alternative splicing during reactivation [25 , 27 , 34] . Adding to the complexity of this transcriptional unit during reactivation , we observe substantial backsplicing from exon 8 to exon 2 in reactivated Akata cells , with lower abundance backsplicing from exon 7 to exon 2 ( Figs 7 and 2 ) . While the extent of reactivation and expression of LMP2 in anti-IgM treated Mutu I cells is lower than that observed in reactivated Akata cells , LMP2 exon 8 to exon 2 backsplicing is also detected in this cell model ( Figs 7 and 2 ) . In contrast , the type III cell lines Jijoye and JY which constitutively express LMP2A show both a lack of alternative forward splicing and circLMP2_E8_E2 backsplicing suggesting context specific regulation of circLMP2_E8_E2 backsplicing ( although some backsplicing from exons 5 to 3 and at exon 6 is observed in JY cells and exon 8 to 2 backsplicing is observed in IB4 cells ) ( Fig 7 ) . This data indicates that circular RNA formation at the LMP2 locus is context dependent , raising the possibility that tissue specific splicing factors may control the levels of backsplicing through cis regulatory elements in the primary LMP2 transcript . The first exon of LMP2A encodes a cytoplasmic signaling domain and exons 2 through 8 encodes the multi-pass transmembrane domain , with exon 9 composed entirely of 3’ untranslated region ( UTR ) sequences [35] . A second form of LMP2 , LMP2B , utilizes an alternative first exon located between LMP2A exon 1 and 2 [35] . With the first exon of LMP2B being composed entirely of 5’ UTR sequences , translation initiation of LMP2B occurs two bases into exon 2 , giving rise to a dominant negative transmembrane inhibitor of LMP2A encoded by exons 2 through 8 [36] . Analysis of coverage and forward splicing data from RNase R-seq from reactivated Akata cells shows coverage and sequential forward splicing across each of exons 2 through 8 ( with albeit low but detectable levels of alternative splicing ) ( Fig 8A ) suggesting that the major circLMP2_E8_E2 species includes all coding sequences of LMP2B . To further explore the potential canonical forward spliced structure of circLMP2_E8_E2 , we performed RT-PCR in reactivated Akata cells using a reverse primer in exon 2 and a panel of forward primers extending from each of exons 4–7 . Each of these primer pairs gave rise to primary PCR fragments consistent with consecutive forward splicing across each of these exons ( Fig 8B ) which was confirmed by sequencing the excised bands . This indicates that the predominant structure of circLMP2_E8_E2 likely contains all of the coding sequences of the dominant negative LMP2B transcript . While the identification of possible function of circLMP2_E8_E2 will require further investigation , cellular distribution analysis shows low but significant cytoplasmic localization ( Fig 8C ) , leaving open the possibility that this circular form of LMP2 could potentially represent a novel isoform of LMP2B . At the same time , analysis of previously published PAR-CLIP data from seven different EBV positive cell lines including lymphomas and the nasopharyngeal carcinoma cell line , C666 [37–40] did not reveal evidence of microRNA association with LMP2 exons 2–8 suggesting that circLMP2_E8_E2 may not play a microRNA sponge function . BHLF1 is a single exon lytic gene that is expressed at low levels during latency but is induced to high levels during reactivation . In rough accordance with the expression profile of the parental BHLF1 gene , BHLF1 backsplicing was detected in most latency conditions and was found to be one of the most highly detected backsplices in the cell during reactivation ( Figs 2 and 9 and S3 Fig ) . Using divergent primers to amplify across the backsplice junction , RNase R resistant backsplices were detected at high levels in reactivated Akata and Mutu cells and at lower levels in Jijoye , JY , SNU719 and YCCEL1 ( Fig 9B ) . Sequencing of multiple PCR fragment clones from these amplified bands revealed not only the backsplice junction identified by find_circ but also a second , more frequently detected BHLF1 backsplice isoform derived from a non-canonical splice donor located 9 base pairs upstream from the circBHLF1 splice donor . During in silico validation of backsplicing ( i . e . aligning sequence reads to the conjoined BHLF1 backsplices ( mentioned above ) ) , we also noted a high number of reads that similarly contained this 9 base splice donor shift ( read counts not included in analysis of circBHLF1 in Fig 2 or S3 Fig ) . Realignment to conjoined backsplice junction sequences for this “alternate” backslice junction yielded 7310 junction spanning reads in reactived Akata cells , 3-fold more than the circBHLF1 backsplice reads in these conditions . To uniquely assess subcellular localization of each of these isoforms , we designed additional divergent primers with forward primers spanning either of the circBHLF1 and “circBHLF1-alt” backsplice junctions and a common reverse primer . Using these primers , the appropriate PCR fragments were amplified and circBHLF1 and circBHLF1alt were both found to display RNase R resistance ( S4 Fig ) . Using these primers for RT-qPCR , both circular BHLF1 RNAs display predominantly nuclear localization ( Fig 9C ) . Further , using BaseScope technology to specifically visualize localization of juxtaposed circBHLF1 backsplice junctions , predominantly nuclear localization was observed ( Fig 9D ) . Together , this data indicates that these BHLF1 derived circular RNAs are abundant during reactivation , resistant to RNase R , and localize to the nucleus , suggesting possible nuclear functions . It is notable that circBHLF1 is located proximal to the lytic origin of replication ( Fig 9A ) . Rennekamp and Lieberman [41] showed that a BHLF1 RNA associates with oriLyt DNA sequences through an R-loop configuration to facilitate lytic viral DNA replication . With the nuclear localization of circBHLF1 and circBHLF1alt and the cis proximity of their originating locus to oriLyt ( Fig 9A ) , worth considering is the possibility that their closed circular structure could provide unique molecular features that engage with the replication complex and regulate lytic viral DNA replication .
Despite more than 50 years of study since the discovery of EBV , our appreciation for the extent and diversity of the EBV transcriptome has grown substantially over the past 15 years with findings of hundreds of new viral RNAs including EBV encoded microRNAs [42 , 43] , a viral ( v ) -snoRNA [4] , stable intronic sequence ( sis ) RNAs [3] and scores of previously unknown polyadenylated and non-polyadenylated lytic transcripts [25 , 27 , 28 , 34 , 44] . Here we report that EBV expresses a repertoire of yet another class of RNAs , circRNAs , in both latency and reactivation . Many of the EBV encoded circRNAs are expressed at levels that are comparable to or higher than the majority of cellular encoded circRNAs ( Fig 2B ) , supporting the contention of potential functional relevance . Some EBV circRNAs are expressed broadly across latency types ( e . g . circRPMS1_E4_E3a , circRPMS1_E4_E2 , and circEBNA_U ) suggesting roles in fundamental processes during latent infection . Further , circRPMS1_E4_E3a and circRPMS1_E4_E2 were found to be expressed in two of two EBV positive stomach cancer biopsies tested , supporting in vivo relevance and possible roles in supporting the tumor phenotype . Though expressed during latency , the expression of circEBNA_W1_C1 and circEBNA_W2_C1 is restricted to type III latency and could be involved in type III latency specific processes such as facilitating Cp initiated transcript diversity , Cp promoter regulation , or other as yet unappreciated type III latency functions . circBHLF1 is detected in most latency cell models but displays extraordinarily high expression under reactivation conditions ( S3 Fig ) . The linear form of BHLF1 is expressed at low levels during latency through the activity of an alternative latency promoter [45 , 46] . Whether the observed expression of circBHLF1 in latency cell lines is due to transcription from the latency promoter and/or whether its expression derives from a small percentage of spontaneously reactivating cells is unclear at this time . Nevertheless , its high expression during reactivation , its proximity to OriLyt , and the known association of BHLF1 RNAs with OriLyt supports speculation that it could play a role in reactivation such as facilitating lytic DNA replication . Expression of circLMP2_E8_E2 displayed tissue specificity , being detected in both reactivation models tested but not in the type III latency cell lines , Jijoye and JY ( with low levels detected in IB4 ) . This data supports the involvement of cis and trans mechanisms that facilitate the regulation of LMP2 back-splicing and they support a potential unique role for circLMP2_E8_E2 in reactivation . Together , the findings reported here reveal a spectrum of EBV circRNAs with diverse expression profiles and likely unique roles in latent and lytic infection . While most of the backspice junctions that we detected are located in latency gene loci and while several candidate viral circular RNAs were found to be expressed in the latency setting , it is notable that there are substantially more backsplice junctions detected during reactivation ( Fig 2 ) . During reactivation , there is substantial subversion of the cell transcription machinery by viral transcription leading to a predominance of viral transcripts in the cell [47] . Liang et al [48] showed that cell stress and resulting limiting of the splicing machinery can increase circular RNA formation relative to linear canonical splicing . This raises the possibility that reactivation might similarly stress the splicing machinery and induce circular RNA formation in this setting . An analysis of backsplicing to canonical splicing ratios of cellular genes , however , did not show a substantial induction of circular RNA formation during reactivation ( S5 Fig ) . This suggests that the increase in viral backsplicing observed during reactivation may be due to virus related mechanisms rather than generalized stresses on the cell itself . This doesn’t preclude the possible involvement of local intra-cellular stresses on the splicing machinery during reactivation , however; for example in viral replication factories where there are high local concentrations of transcribing viral genomes . Even in this scenario , though , there must be cis specificity since not all spliced viral transcripts show backsplicing during reactivation ( e . g . BZLF1 , see S6 Fig ) . This supports the contention that backsplicing of viral genes during reactivation is regulated and that the induced expression of viral circular RNAs may have functional relevance . Notably , the lytic EBV BMLF1 protein has been shown to interact with splicing factors and to be involved in RNA transport , polyadenylation , splicing , and translation [49–54] . It will be important to determine whether BMLF1 plays a role in enhancing and/or regulating lytic backsplicing and viral circRNA formation . Further , with its known function in enhancing translation of some mRNAs , it will be interesting to assess whether BMLF1 potentially facilitates translation of cytoplasmically localized circRNAs such as circLMP2_E8_E2s . One of the most notable functions of cellular circRNAs is the regulation of microRNA activity through sponging/sequestration mechanisms . Nevertheless , mining previously published PAR-CLIP data from seven EBV positive cell line models [37–40] , we were unable to identify interactions between any viral or cellular microRNAs and EBV sequences spanning circular RNA exons . Since all of these studies were performed in latent cell systems , this does not preclude possible interactions between microRNAs and lytic EBV circRNAs , but does support the contention that the EBV circRNAs identified here serve other non-microRNA sponge functions . While circRNAs are thought to mostly function through non-coding mechanisms , some have been shown to localize to ribosomes and to be translated [22–24] . Bencun et al [55] have previously performed ribosomal profiling of B-lymphocytes infected with the B95-8 and M81 strains of EBV to assess the translation of latent ( B95-8 infection ) and lytic ( M81 infection ) viral transcripts . Although circEBNA_U contains only a single 6-amino acid open reading frame ( ORF ) , cells treated with harringtonine to map translation initiation sites showed a peak covering this short ORF [55] . Further , as noted in [55] , the EBNA U exon contains a previously identified IRES that overlays this short ORF that is thought to regulate translation of downstream spliced EBNA reading frames [56] . In the circEBNA_U context , the IRES could conceivably self-regulate translation initiation of this ORF to generate a short functional peptide or to mediate some other regulatory function through an association with ribosomes . Notably , circBHLF1 has the potential to code for a 200 amino acid ORF through two consecutive frame shifts at the backsplice junction ( S7 Fig ) . Alignment of harringtonine/ribosomal RNA-seq data from B95-8 or M81 infected cells showed weak evidence of translation initiation at this site [55] . Although our RT-qPCR analysis showed nearly exclusive nuclear localization of circBHLF1 , given the abundance of circBHLF1 , it is possible that low levels of circBHLF1 could potentially associated with ribosomes and generate a translated product from this ORF . While we observed minor harringtonine/ribosomal RNA-seq peaks at the initiation codon located at the beginning of exon 2 of LMP2 , in the context of these infections , circLMP2 may not be expressed and/or any initiation at this site could represent initiation from linear LMP2B transcripts derived from the LMP2B promoter . Assessing whether circLMP2_E8_E2 is a novel isoform of LMP2B will require more detailed experiments to specifically assess circLMP2_E8_E2 association with ribosomes and perhaps epitope tagging of circLMP2_E8_E2 . Given the diverse spectrum of latency and lytic genes from which EBV circRNAs originate , their diverse patterns of expression and their different sub-cellular distributions , EBV circRNAs may play roles in a wide array of latent , lytic , nuclear and cytoplasmic functions in latency , reactivation , B-cells and epithelial cells in natural EBV infection and potentially in EBV-associated malignancies . Together , this study uncovers a new spectrum of virus encoded RNAs and should set the stage for new lines of research into virus biology and EBV-associated lymphoma and epithelial cancers . In particular , the finding of latency associated circRNAs adds to the repertoire of potential future therapeutic targets . Further , with their increased stability compared to linear RNAs , circRNAs are receiving attention as potential liquid biopsy markers [20 , 57] . With antibodies to EBV antigens previously identified as predictors of nasopharyngeal carcinoma ( e . g . [58–60] ) , serum viral circRNAs could someday show predictive and/or diagnostic potential in EBV associated disorders .
Akata ( obtained from Kenzo Takada ) , Mutu I , Sav I , Mutu III , IB4 , Jijoye , JY ( Mutu I , Sav I , Mutu III , IB4 , Jijoye , and JY cell lines obtained from the laboratory of Samuel H Speck ) and SNU719 ( Korean Cell Line Bank ) cells were cultured in RPMI 1640 media ( Fisher Scientific , catalog no . SH30027 ) plus 10% fetal bovine serum ( FBS ) ( Thermo Fisher , catalog no . 10437 ) . YCCEL-1 ( Korean Cell Line Bank ) cells were grown in Eagle's minimum essential medium ( EMEM ) ( ATCC , catalog no . 30–2003 ) supplemented with 10% FBS . All cells were cultured at 37°C in a 5% CO2 incubator . Akata and Mutu I cells were spun down and resuspended at a concentration of 106 cells/ml in fresh RPMI 1640 medium ( 10%FBS ) . Anti-human IgG ( Sigma-Aldrich , catalog no . I5260 ) or anti-human IgM ( Sigma-Aldrich , catalog no . I0759 ) was added to Akata and Mutu I cell suspensions , respectively , to a final concentration of 10 ug/ml . Treated and untreated cells were harvested 24 h and 48 h later for Akata and Mutu I cells , respectively , and subjected to RNA isolation . Whole cell RNA preparations were carried out using TRIzol reagent ( Thermo Fisher , catalog no . 15596 ) according to the vendor’s recommended protocol . For tumor and normal tissue , pieces were first ground finely using a mortar and pestle in liquid nitrogen prior to disruption with TRIzol reagent . Nuclear and cytoplasmic RNAs were isolated using the Cytoplasmic & Nuclear RNA Purification Kit from Norgen Biotek Corp . ( catalog no . 21000 ) according to the vendor’s protocol . All RNA preparations were subjected to DNase treatment twice using the DNA-free kit ( Thermo Fisher , catalog no . AM1906 ) . RNA-sequencing was performed at the Beijing Genomics Institute ( BGI ) . For polyA-seq , RNAs were selected using a poly dT column , and for ribodepletion-seq ribosomal RNAs were excluded using hybrid capture for 28S , 18S and 5S ribosomal RNAs . For RNase R-seq , RNAs were subjected to DNA and rRNA depletion , followed by linear RNA depletion using RNase R . For all sequencing , Truseq stranded libraries were generated and sequenced using 2x100 base sequencing on a HiSeq 4000 system . Back-splicing was analyzed using find_circ [26] with default parameters . For in silico validations , a STAR [61] genome index was generated which contains the human hg38 genome build plus conjoined backsplice junctions for each candidate EBV backsplice identified by find_circ . Raw fastq files were aligned to this combined genome index using STAR ( –outFilterMultimapNmax 20 –outSAMtype BAM SortedByCoordinate–outWigType wiggle–outWigNorm None ) and reads spanning EBV backsplice junctions with a minimum of 12 base overlap ( minimum of 90% homology ) on each side of the junction were pulled out for visualization on the Integrative Genomics Viewer ( IGV ) and number of reads mapping to each junction in each cell line were quantified for reporting . Notably , while intron lariats are similarly covalently closed RNAs and can be enriched by RNase R treatment , lariat RNAs are typically unstable due to the action of debranching endonucleases that target the 5’ to 2’ junction . In addition , reverse transcriptase displays resistance crossing the 5’ to 2’ junction , thereby further under-representing lariat junction reads in RNA-seq data . Further , all EBV backsplice junctions occurred at canonical GU AG splice donor-acceptor motifs and showed no evidence of micro–insertions/-deletions or single nucleotide substitutions that are typically observed following reverse transcription across lariat junctions . We therefore conclude that all 33 EBV backsplice junctions reported here likely represent true circRNAs . For canonical splicing and coverage display , RNA-seq data from polyA libraries were analyzed by STAR alignment against the human hg38 plus chrEBV_Akata_inverted genome using STAR ( –outFilterMultimapNmax 20 –outSAMtype BAM SortedByCoordinate–outWigType wiggle–outWigNorm None ) . Splice junction data from . SJ files and wiggle output files were used to generate junction read numbers and coverage information . Backsplice read counts were extracted from . bed junction count files derived from find_circ output . For visualization of forward splicing , coverage and backsplicing , in house software ( circleVis ) was developed . Exon level coverage was represented by color intensity with canonical and backsplice junction curves plotted above and below the exon diagram , respectively . Each junction count was individually plotted . For visualization of splicing and coverage across the EBNA locus , a custom analysis pipeline was developed to accommodate unique problems associated with aligning to repetitive elements . Specifically , all canonical forward splice junctions were quantified from STAR alignments to an artificial mini EBV genome containing upstream unique sequences , two copies of BamHI W repeats , and downstream unique regions extending past the EBNA1 gene . Coverage and backsplice information was generated using STAR ( coverage ) and find_circ ( backsplice counts ) alignments to an artificial mini EBV genome containing one copy of the BamHI W repeats inserted between unique upstream and downstream EBNA locus sequences . For display , forward spliced read counts , coverage and backsplice read counts associated with the W1 and W2 exons were divided by the appropriate number of exons/junctions to distribute these counts equivalently across 7 W repeats . 5 ug of total RNA was incubated with or without 20 units of RNase R ( Applied Biological Materials , Inc . , catalog no . E049 ) . Briefly , no ( control ) or 1 . 5 ul RNase R ( 30u ) ( test ) and 3 ul of 10X RNase R buffer were added to 5 ug of RNA in a total volume of 30 ul and incubated in a 37C water bath for 30 minutes . Either no ( control ) or 1 . 5 ul ( 30 units ) ( test ) more RNase R was added to reactions and incubated for an additional 1 . 5 hours in a 37C water bath . RNAs were cleaned and concentrated using the RNA Clean & Concentrator-5 kit ( Zymo Research , catalog no . R1015 ) and eluted in 10ul H2O for use in PCR reactions . cDNA was synthesized from total RNA ( control or RNase R treated where indicated ) using SuperScript IV First-strand Synthesis System ( Thermo Fisher , catalog no . 18091 ) and the cDNAs were amplified by taq-PCR ( Thermo Fisher , catalog no . 11304 ) following the vendor’s protocol . PCR products were run on a 1 . 5% agarose gel at 4°C . PCR products were cut out and purified using the NucleoSpin Gel & PCR Clean-up Kit ( Clontech , catalog no . 740609 ) . The resulting PCR fragments were cloned into the pCR4-TOPO vector ( Thermo Fisher , catalog no . 450030 ) and the inserts were Sanger sequenced . For assessing nuclear/cytoplasmic localization , cDNA was synthesized from total RNA using SuperScript IV First-strand Synthesis System ( Thermo Fisher , catalog no . 18091 ) according to the manufacturer’s protocol . qPCR analysis was performed using iQ SYBR Green Supermix ( Bio-Rad , catalog no . 170–8882 ) on a Bio-Rad CFX96 instrument as follows: 1 μl of cDNA and 1 μl of 10 μM primers were mixed with 10 μl of SYBR green supermix and 8 μl nuclease-free H2O to a 20 μl reaction volume . Polymerase was activated and cDNA was denatured at 95°C for 5 minutes . cDNA was then amplified for 40 cycles with 15 s denaturation at 95°C , 60s annealing/extension and plate reading at 60°C . Melting curve analysis was performed at temperatures from 60°C to 90°C with 0 . 5°C increment per 5 s . Expression fold changes were calculated using the comparative CT method ( 2-ΔΔCT ) . Cytoplasmic enrichment was calculated as 2^ ( ( ( test gene ( Ctnuclear—Ctcytoplasmic ) ) – ( nuclear reference gene ( Ctnuclear—Ctcytoplasmic ) ) ) / ( ( cytoplasmic reference gene ( Ctnuclear—Ctcytoplasmic ) ) – ( nuclear reference gene ( Ctnuclear—Ctcytoplasmic ) ) ) . ACTB forward: 5’- CGTCATACTCCTGCTTGCTG ACTB reverse: 5’- GGCATCCTCACCCTGAAGTA ACTB ( qPCR ) forward: 5’- CACTCTTCCAGCCTTCCTTC ACTB ( qPCR ) reverse: 5’- GTACAGGTCTTTGCGGATGT GAPDH exon 1 forward: 5’- AAGGTGAAGGTCGGAGTCAAC GAPDH exon 2 reverse: 5’- GGGTCATTGATGGCAAC KCNQ1OT1 forward: 5’- TACCGGATCCAGGTTTGCAGTACA KCNQ1OT1 reverse: 5’- GCTGATAAAGGCACCGGAAGGAAA circEBNA exon W1 ( Akata ) forward: 5’- GGGAGACCGAAGTGAAGTCC circEBNA exon W1 ( B95-8 ) forward: 5’- GGGAGACCGAAGTGAAGGCC circEBNA exon C1 ( W1 ) reverse: 5’- ATGGTAAGAACCCTGCGATG circEBNA exon W1-C1-sp forward 5’- TATCGGGCCAGAGATGGCAT circEBNA exon C1 reverse ( W1-C1-sp ) 5’- TAGATGATTTGCGGGTTACATGA circEBNA exon W2 forward: 5’- AGAACCCAGACGAGTCCGTA circEBNA exon C1 ( W2 ) reverse: 5’- CATGGTAAGAACCCTGCGAT circRPMS1 exon 4 forward: 5’- CTAGTGCTGCATGGGCTCCT circRPMS1 exon 3a reverse: 5’- GTCATACGCCCGTATTCACA circRPMS1 exon 4-3a-sp forward: 5’-GCTGTTCCTGAACGACGAG circRPMS1 exon 4-3a-sp reverse: 5’- ACACGCCGGACCTTGCC circBHLF1 forward: 5’- CCAGAGGAGCCCCAGAAC circBHLF1var-sp forward: 5’- CCCAGAACCAGGTGCACC circBHLF1-sp forward: 5’- AGGCAAGCCGGGTGCAC circBHLF1 reverse: 5’- ATGCTGCATCCGCTAGTCC LMP2A exon 1 forward: 5’- CTACTCTCCACGGGATGACTC LMP2A exon 2 reverse: 5’- AGGTAGGGCGCAACAATTAC circLMP2 exon 4 forward: 5’- TTCTGGTGATGCTTGTGCTC circLMP2 exon 5 forward: 5’- TCACTGATTTTGGGCACACTT circLMP2 exon 6 forward: 5’- ATCGCTGGTGGCAGTATTTT circLMP2 exon 7 forward: 5’- GCTCTCGCACTCTTGTTGCT circLMP2 exon 8 forward: 5’- TCATTAGATGCTGCCGCTAC circLMP2 exon 2 reverse: 5’- AGGTAGGGCGCAACAATTAC To assess expression across cell lines , cDNA was synthesized using SuperScript IV First-strand Synthesis System ( Thermo Fisher , catalog no . 18091 ) according to the manufacturer’s protocol . qPCR analysis was performed using TaqMan Fast Advanced Master Mix ( Thermo Fisher , catalog no . 4444557 ) on a Bio-Rad CFX96 instrument as follows: 1 μl cDNA and 1 μl 20X TaqMan assay were mixed with 10 μl of 2X TaqMan Fast Advanced Master Mix and 8 μl nuclease-free H2O to a 20 μl reaction volume . After a 2 minute UNG incubation at 50°C , polymerase was activated at 95°C for 2 minutes . cDNA was then denatured and amplified/extended for 40 cycles with a 3 s denaturation at 95°C and a 30 s annealing/extension and plate reading at 60°C or 66°C . The TaqMan assays were designed using the following primers , with the probe primers designed to hybridize to 13 bases on each side of the backsplice junction: circRPMS1_E4_E3a: circRPMS1 exon 4 forward: 5’-CTAGTGCTGCATGGGCTCCT circRPMS1 exon 3a reverse: 5’-GTCATACGCCCGTATTCACA circRPMS1_E4_E3a ( Probe ( fluorescein ) ) : 5’-gtcgacgggcaaGGtccggcgtgtcc BHLF1: circBHLF1 forward: 5’- CCAGAGGAGCCCCAGAAC circBHLF1 reverse: 5’- ATGCTGCATCCGCTAGTCC circBHLF1 ( Probe ( fluorescein ) ) : 5’-ccaggcaagccgGGtgcaccggacca EBNA-U: circEBNA_U forward: 5’-TCTGGAGCCTGACCTGTGA circEBNA_U reverse: 5’-TCTCTGGGCTGCAGAATCA circEBNA_U Probe ( fluorescein ) : 5’-cgtctcctttaaGGtaacttaggaag RPL30 ( reference gene ) : RPL30 forward: 5’-CACCAGTTTTAGCCAACATAGC RPL30 reverse: 5’-TGAAGATGATCAGACAAGGCAA RPL30 ( Probe ( fluorescein ) ) : 5’-TTCTCGCTAACAACTGCCCAGCT Relative expression was calculated using the delta delta Ct method using RPL30 as the reference gene and uninduced Akata cells as the reference sample . Cell pellet preparation was carried out in accordance with the vendor’s protocol ( Advanced Cell Diagnostics ) . Specifically , Akata cells were cultured with or without 10 μg/ml of anti-human IgG ( Sigma-Aldrich , cat # I5260 ) . Twenty-four hours post treatment , cells were harvested and washed once with 1X phosphate-buffered saline ( PBS ) . Cell pellets were resuspended in 10ml of 10% neutral buffered formalin ( Sigma-Aldrich , Cat # HT501128 ) with 40 ml 1X PBS and fixed at room temperature for 24 h . Cells were washed twice with 10ml of 1X PBS and the cell pellets were resuspended in liquified HistoGel ( ThermoFisher , Cat # HG-4000-012 ) and solidified on ice . Cell blocks were then paraffin embedded and 5 μm sections were mounted onto Superfrost plus slides ( Fisher , Cat # 4951PLUS ) . BaseScope assays were performed in accordance with the BaseScope Reagent Kit-RED protocol ( Advanced Cell Diagnostics , Cat # 322970 ) using a circBHLF1 backsplice specific probe ( Advanced Cell Diagnostics , Cat # BA-V-EBV-BHLF1-circRNA ) . Assays were performed at room temperature unless otherwise indicated . Sections were baked at 60°C for 1 h , followed by deparaffinizing in xylene twice for 5 min , ethanol twice for 5 min , and baked at 60°C for 5 min . Slides were first treated with hydrogen peroxide for 10 min at RT , treated with target retrieval reagent for 15 min at 100°C , and treated with protease III for 15 min at 40°C , with two distilled water rinses between each treatment . BaseScope probes were then applied and incubated for 2 h at 40°C in a HybEZ oven followed by incubation with reagents AMP0 for 30 min at 40°C , AMP1 for 15 min at 40°C , AMP2 for 30 min at 40°C , AMP3 for 30 min at 40°C , AMP4 for 15 min at 40°C , AMP5 for 30 min at RT and AMP6 for 15 min at RT . Slides were rinsed twice with wash buffer for 2 min between each AMP incubation . Slides were incubated with Fast Red for 10 min at RT in the dark , then counterstained with Gill’s hematoxylin ( Sigma-Aldrich , Cat # GHS132 ) for 15 min at 60°C and mounted in ProLong Gold Antifade Mountant with DAPI ( ThermoFisher Scientific , Cat # P36941 ) . The BaseScope Fast Red and DAPI signals were visualized on an Eclipse Ti2 inverted microscope system ( Nikon ) . | Our understanding of the extent of EBV transcriptome complexity has recently come to light with discoveries of viral microRNAs , snoRNAs , and a diverse set of transcript isoforms expressed during reactivation . While EBV utilizes distinct transcriptional programs during it’s normal infection cycle , the latency programs are critical drivers of EBV associated oncogenesis where they activate oncogenic pathways without killing the cell via lytic viral replication . Further , pathogenic latency programs typically express minimal levels of viral protein coding genes and instead rely on non-coding RNA mechanisms to alter the host cell environment . The utilization of non-coding RNAs in these settings is presumably an effort to minimize infected cell killing by the immune system through adaptive immune responses . Using tissue culture models representing the spectrum of EBV transcriptional states that occur in vivo , we report here that EBV expresses the circular class of predominantly non-coding RNAs . This stands to reveal a previously unknown set of players in EBV associated biology and disease . The findings reported here set a key platform from which future functional and potentially clinical related studies can be carried out to bring the initial findings from discovery to a new understanding of EBV biology and disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"nucleases",
"enzymes",
"pathogens",
"gene",
"regulation",
"dna-binding",
"proteins",
"enzymology",
"microbiology",
"viruses",
"micrornas",
"dna",
"viruses",... | 2018 | The Epstein Barr virus circRNAome |
A purely information theory-guided approach to quantitatively characterize protease specificity is established . We calculate an entropy value for each protease subpocket based on sequences of cleaved substrates extracted from the MEROPS database . We compare our results with known subpocket specificity profiles for individual proteases and protease groups ( e . g . serine proteases , metallo proteases ) and reflect them quantitatively . Summation of subpocket-wise cleavage entropy contributions yields a measure for overall protease substrate specificity . This total cleavage entropy allows ranking of different proteases with respect to their specificity , separating unspecific digestive enzymes showing high total cleavage entropy from specific proteases involved in signaling cascades . The development of a quantitative cleavage entropy score allows an unbiased comparison of subpocket-wise and overall protease specificity . Thus , it enables assessment of relative importance of physicochemical and structural descriptors in protease recognition . We present an exemplary application of cleavage entropy in tracing substrate specificity in protease evolution . This highlights the wide range of substrate promiscuity within homologue proteases and hence the heavy impact of a limited number of mutations on individual substrate specificity .
Proteases catalyze cleavage of peptide bonds and are involved in virtually all fundamental cellular processes [1] turning proteases into central drug targets [2] . Far over 500 proteases with unique substrate cleavage patterns have been identified in the human genome [3] . These patterns reach from specificity for a single peptide to broad spectra of cleaved peptides . For instance , digestive enzymes are known to process a wide range of substrate sequences in contrast to proteases involved in signaling pathways cleaving only very distinct peptide bonds [1] . These signaling cascades include the blood-clotting cascade [4] , apoptosis pathways [5] and regulatory activation steps of digestive proteases [6] . Specificity of a protease is determined by interactions in the protein-protein interface of protease and substrate . The spectrum of substrates to be cleaved is classified by subpocket-wise interactions following the convention of Schechter and Berger [7]: The peptide's scissile bond is designated between N-terminal P1 and C-terminal P1′ . These subpocket indices are incremented over sequential amino acids . Protease interface residues are numbered accordingly over all subpockets Sn-Sn′ , thus ensuring that interacting residues are indexed with the same number . Binding modes of processed polypeptides are highly similar due to the fact that the substrate is locked in an extended beta conformation within the protease binding site [8] , [9] . This canonical conformation usually includes residues in the P3-P3′ substrate region , at most extended to P5 , in serine protease elastase [10] . Cleavage specificity is generally originating from distinct molecular interactions between substrate and enzyme . Simple cleavage rules for serine proteases only rely on the prominent P1-S1 interactions . For instance , the hydrophobic S1 pocket of chymotrypsin causes specificity for substrates providing hydrophobic residues at their P1 position . In contrast , an Asp residue in the S1 site of the homologous trypsin determines specificity for Arg and Lys at P1 [11] . Limitations of such simple models are evident , as S1-directed mutation does not allow transposition of trypsin specificity to chymotrypsin [12] . Moreover , complex adjacent protein-loop interactions and dynamics were found to determine substrate specificity [13] , [14] . Interactions between enzyme and substrate span several subpockets in the protease binding site . Experimental data shows that S2–S3 sites hardly affect substrate specificity in chymotrypsin [15] , but account for specificity of the homologous elastase [16] . Especially chymotrypsin-like enteropeptidase shows exceptional specificity in the S5-S1-region cleaving only substrates containing the sequence Asp-Asp-Asp-Asp-Lys as trypsinogen [17] . P4-S4 interactions are found to be highly specific in case of the non-homologous subtilisin serine proteases [18] . Especially in the S1-S4-region , closely homologous serine proteases show significant differences in respective cleavage specificity reaching from limited proteolysis to almost unspecific substrate cleavage . Several cleavage site prediction tools are based on such simple and intuitive rules and are available online [19] . A plethora of experimental cleavage data for proteases is available in several databases . Cleavage information is generated experimentally by several methods reviewed by Diamond [20] and Poreba and Drag [21] reaching from fluorescence-based assays [22] , isotopic labeling techniques [23] , biotinylation schemes [24] over phage display [25] , library-based approaches [26] , microarray-based methods [27] , [28] and combinations thereof to modern high-throughput techniques as proteomic identification of cleavage sites ( PICS ) [29] , [30] . Cleavage data is accessible in several public databases including the MEROPS database [31] , [32] linking structural protease data to cleavage activity . Although cleavage information for known proteases is easily accessible , by now no attempt has been made to develop a quantitative measure for subpocket-wise and total protease specificity in contrast to pure feature extraction techniques as for example cascade detection [33] . Analysis of protease cleavage data was mostly limited to qualitative interpretation by conversion into consensus recognition motives and visualization by sequence logos [34] , iceLogo [35] or heat maps [29] . We propose the usage of information entropy to merge experimental cleavage data into an easily interpretable score for subpocket specificity as well as overall protease specificity . Following the idea of information entropy [36] , which is consistent with entropy in statistical mechanics [37] , we developed an information theory-based specificity score named “cleavage entropy” . These cleavage entropy values depict a measure for uncertainty , and hence strictness of substrate readout , directly related to the information content of each amino acid position in a cleavage motif . A similar approach was successfully applied for description of sequence specificity of DNA binding proteins [38] and substrate promiscuity of whole enzyme families [39] , including the P-region of proteases as an example [40] . DuVerle and Mamitsuka used information entropy for selection of a set of proteases showing diverse cleavage patterns and hence substrate promiscuity [41] .
To generate subpocket-wise specificity entropies , cleavage data were extracted from the MEROPS database [31] . Comparable cleavage databases as the CutDB [42] or Proteolysis MAP [43] were found to provide less cleavage information . Proteases of diverse families containing at least 100 substrate entries form a data set of 47 proteases . Methionyl aminopeptidases were excluded from the analysis , as positions P4-P2 remain unoccupied by the substrate upon cotranslational removal of N-terminal methionine residues . A complete sequence matrix containing the absolute occurrence of 20 amino acids at eight subpockets P4′ to P4 was compiled for each protease . Protease-wise cleavage sequence matrices were normalized according to the natural abundance of individual amino acids [44] . Subsequently , a second normalization to 1 at each subpocket yielded a data matrix containing probabilities for each substrate amino acid at each protease subpocket . Information theory-based cleavage entropy is defined according to Formula 1 taking into account the whole distribution of amino acids at each position rather than a single peak of elevated amino acid abundance . Substrate information is purely incorporated as sequence , not covering any kind of secondary structure information . Derived dimensionless subpocket-wise entropy values , measure the broadness of distribution of cleaved substrates , range from 0 for a perfectly conserved single amino acid to 1 for an equal distribution of substrates , reflecting complete unspecific substrate binding . Formula 1: Calculation of subpocket-wise cleavage entropy Si from subpocket-wise amino acid probabilities in known substrates pa , i . Subpocket-wise substrate specificity information is of high interest to compare individual subpockets of a single protease and individual specifity profiles between proteases . To facilitate analysis of different proteases as a whole , a summation of individual subpocket cleavage entropies yields quantitative overall cleavage entropy per protease ( see Formula 2 ) . This total cleavage entropy over eight substrate positions in the central binding site region ( P4 to P4′ ) allows for ranking of proteases with respect to their whole substrate specificities . Entropy values range from 0 for a single conserved substrate to 8 for a random distribution of amino acids in cleaved substrates . Formula 2: Calculation of overall protease cleavage SCleavage entropy by summation of 8 subpocket-wise cleavage entropies Si from P4-P4′ subpockets . Although cooperativity effects between subpockets were described for subtilisins [45] and reviewed by Ng et al . [46] , available cleavage data only allows for a rough estimation of these correlation effects besides independent study of subpocket specificity . To cover inter-subpocket correlation effects in detail , data simply based on known substrates is too sparse . An extension from purely qualitative cleavage information to substrate-dependent quantitative binding affinity or kinetics measurements would be necessary . A suitable database containing diverse protease substrates is currently not known to the authors , but could also be of high interest to weight individual substrate contributions in order to refine the current implementation . A smaller set of fluorescence-based substrate turnover measurements for proteases was published by Harris et al . [22] , but is restricted to variation of the P-region in substrates for eight proteases . As only trypsin provides a sufficient data basis to study subpocket correlation effects with more than 14000 substrates listed in MEROPS , we performed an inter-subpocket correlation analysis only for this protease . The one-dimensional subpocket-wise cleavage entropy calculations presented above can directly be extended to a more-dimensional case yielding for two dimensions a pairwise cleavage entropy score depending on amino acids a and b at position i and j and their respective probabilities pa , i , pb , j . Formula 3: Calculation of pairwise cleavage entropy Si , j from subpocket-wise amino acid pair probabilities in known substrates pa , i , pb , j . This measure for inter-subpocket correlation effects yields as in the independent analysis ( cleavage entropy ) a score of 0 for a conserved single amino acid pair and a value of 1 for a distribution of amino acid pairs as expected by random chance from natural abundance [44] . To avoid artifacts from a lacking data basis we set a stringent cutoff of 10000 substrates in this two-dimensional analysis to allow for the same statistics as in the one-dimensional case ( 100 substrates ) . As part of the discussion , protease specificity is compared to evolutionary distance . Sequences downloaded from Uniprot [47] as indexed in the MEROPS database [31] were grouped into respective protease clans . Sequences of each clan were sorted according to total cleavage entropy and aligned by ClustalW using default settings [48] . Tools from the EMBOSS server [49] were used for phylogenetic tree construction: fprotdist using default settings to calculate protein distance matrices , fkitsch using default settings to construct phylogenetic trees using the Fitch-Margoliash method [50] . Phylogenetic trees were visualized using Interactive Tree of Life ( ITOL ) [51] . Protein structure visualizations were created with PyMOL [52] based on the X-ray structures of trypsin and thrombin in complex with BIBR1109 ( PDB: 1G32 , 1G36 ) [53] . A subpocket definition derived from Bode et al . [54] was used for mapping of subpocket-wise cleavage entropies to the binding site region .
Entries with more than 100 annotated substrates in the MEROPS database represent 47 proteases comprise all major protease catalytic types . The three major protease catalytic types , serine , metallo and cysteine proteinases , covering more than 90% of known proteases [9] , represent 40 entries or 85% of the test set . Calculated subpocket-wise cleavage entropies will be discussed by catalytic type to enable comparison of relative variation of binding specificity . Relative importance of subsites in determining cleavage specificity is highlighted by lowered entropy values providing specificity profiles for individual proteases . Serine proteases show pronounced specificity at the P1 substrate site occupying the characteristic deep S1 pocket with an averaged cleavage entropy as low as SP1 = 0 . 256 ( see Figure 1 ) . The low P1 cleavage entropy value reflects widely accepted specificity rules for serine proteases solely based on P1-S1 interactions . A second hotspot for specific interactions of serine proteases is found in the P2-region with an average cleavage entropy of SP2 = 0 . 781 , which is especially lowered for proprotein processing proteases kexin , furin and proprotein convertase 2 cleaving at paired basic residues [55] . Overall , serine proteases tend to bind conserved residues in P-region ( average SP4-P1 = 0 . 696 ) rather than the P′-region ( average SP1′-P4′ = 0 . 912 ) in accordance to findings of Page et al . for coagulation proteases as thrombin [56] . See Figure 2 for a detailed comparison of subpocket-wise cleavage entropies mapped to the three-dimensional structure of thrombin and trypsin . All serine proteases in the test set show pronounced specificity in the P1-region , including even so-called unspecific proteases as trypsin binding to highly conserved arginine and lysine residues at the P1 site . An extension of this specific reading frame in both directions of the substrate is observed for example for thrombin and furin , where the latter protease shows extraordinary specificity at the P4 site independent of other specific residues . These lowered entropy values reflect the proposed Arg-Xaa-Lys/Arg-Arg consensus in the P4-P1-region for furin substrates [57] and confirm general specificity rules for P4 specificity of the subtilisin-like clan of serine proteases [18] . Metallo proteases in general show less intense subpocket-wise specificity patterns than serine proteases . Their substrate readout is most pronounced in the P1′ position with an average cleavage entropy of 0 . 703 ( see Figure 3 ) consistent with findings of Overall et al . for the substrate specificity of matrix metallo proteases [58] . Peptidyl-Lys metallo peptidase reads a perfectly conserved lysine residue at P1′ in all 2111 known substrates . However , P1′ is not the most specific subpocket in all metallo proteases . Further subpockets showing less pronounced substrate readout are located at P3 ( SP3 = 0 . 751 ) and P3′ ( SP3′ = 0 . 829 ) in analogy to computational predictions of Pirard [59] . Little substrate specificity is observed for other binding sites leading to an almost equivalent average substrate specificity over the whole P-and P′- region ( SP4-P1 = 0 . 832 , SP1′-P4′ = 0 . 831 ) . We find matrix metallo proteases ( MMPs ) to differ in their substrate specificity from other members of the metallo proteases . Cleavage entropy calculation highlights the P1′ position as major determinant of specificity in MMP-2 , hence named “specificity pocket” [60] , whereas other subsites show little substrate preferences . Additionally , a preference for proline at P3 has been observed [61] , [62] , which is consistent with lowered cleavage entropy values at P3 found throughout the MMP family . MMP-13 shows particular preference for proline residues at P3 reducing cleavage entropy to 0 . 455 . Strikingly , particular metallo proteases span substrate specificity over all covered subsites: The highly specific members thimet oligopeptidase and neurolysin show cleavage entropy values lower than 0 . 850 throughout all subpockets . Cysteine proteases are characterized by cleavage entropies comparable to serine proteasaes rather than metallo proteases . P1 interactions dominate substrate specificity with a cleavage entropy of SP1 = 0 . 630 similar to serine proteases ( see Figure 4 ) . Caspases account for the pronounced P1 interaction in this protease family as well as a smaller second specificity peak at P4 position ( SP4 = 0 . 848 ) . The P-region exhibits most of cysteine protease' substrate specificites with average cleavage entropy SP4-P1 = 0 . 802 compared to the P′-region SP1′-P4′ = 0 . 904 . Caspases are shown to read conserved aspartate residues in P1 position with an extraordinarily high specificity ( P1<0 . 05 ) , a characteristic not present in all other cysteine proteases . Subsite specificity of apoptosis signaling caspases [63] extends over larger areas of the P-region [64] , especially pronounced in case of caspase 7 [29] . In contrast to caspases , calpains cleave broader substrate spectra whilst showing overlap with caspases in some regions of substrate space [65] . Traceable P3′ specificity is only observed for calpains amongst cysteine proteases . Broader distributions of substrates known for cathepsins [66] are quantitatively reflected by higher cleavage entropies . Cathepsin K's subtle substrate specificity at P1 and P1′ ( SP1 = 0 . 680 , SP1′ = 0 . 820 ) has been described by Schilling et al . [29] . Falcipains do not feature any particular subsite specificities , but tend to show complex and promiscuous specificity profiles . Simple counting of cleavage entries would have missed this unspecific behavior , as the number of available cleavage sites annotated in MEROPS is comparably low for falcipains . Besides the three main classes of proteases , six further proteases with more than 100 cleavage patterns were found within MEROPS ( see Figure 5 ) : signal peptidase , containing a rare serine dyad at the active site [67] , forming an active dimer complex in eukaryotes and hence indexed in MEROPS as complex peptidase , as well as five aspartic proteases . Two members of glutamic proteases showing distinct cleavage behavior were added to the sample to include this missing catalytic type , although less known cleaved peptides are indexed . The signal peptidase complex is a membrane-bound protease involved in membrane translocation signaling [68] . Cleavage entropies SP3 = 0 . 726 and SP1 = 0 . 617 reflect the well-established specificity rules for signal peptidases focussing on positions P3 and P1 [69] . Distinct P1 specificity matches classical serine proteases involving a catalytic triad at the active site , whereas P3 readout is not a general characteristic of serine proteases . All five aspartic proteases are found to depend mostly on P1 interactions with an average SP1 = 0 . 768 . Other subpockets in P- and P′-region tend to exhibit likewise unspecific substrate binding ( SP4-P1 = 0 . 892 , SP1′-P4′ = 0 . 909 ) . HIV retropepsin , a prominent target in drug design , shows distinct specificity at P2′ position with SP2′ = 0 . 768 supporting findings of Schilling et al . [29] . Furthermore , specific substrate readout of HIV retropepsin at positions P1 and P1′ was described in the literature [70] and is quantified with lowered cleavage entropies of SP1 = 0 . 792 and SP1′ = 0 . 848 respectively . Aspergilloglutamic and scytalidoglutamic peptidase are added to the data set though sparse cleavage data to cover the group of glutamic peptidases represented by the members with highest number of annotated subtrates ( 68 and 37 respectively ) . Aspergilloglutamic and scytalidoglutamic peptidase provide two examples of variable cleavage profiles amongst the same protease class: Whereas the P1 position shows nearly identically lowered cleavage entropies , scytalidoglutamic peptidase reads substrate residues over the whole range of eight covered subpockets in contrast to aspergilloglutamic peptidase not showing pronounced substrate preferences at other subpockets than P1 . Summing up previous findings , average subpocket cleavage entropy profiles were calculated for protease catalytic types ( see Figure 6 ) . Serine proteases show distinct lowered cleavage entropy at their specific S1 site . Less pronounced S1 specificity is present for cysteine and aspartic proteases , whereas metallo proteases show subpocket cleavage entropy profiles including diverse cleavage entropy minima with the most specific substrate binding in the S1′ site . Summation of subpocket-wise cleavage entropies yields a total estimate of protease specificity ( see Figure 7 ) . The additional information content of calculated total cleavage entropies compared to simple substrate counting is reflected by a squared linear correlation coefficient as low as r2 = 0 . 034 over the core test set of 47 proteases . Likewise , qualitative ranking correlation is comparably low with a Spearman ranking correlation of r = 0 . 334 over 47 proteases . Taking into account the whole distribution of amino acids in known substrates rather than the plain number of known substrates , has the advantage to minimize the impact of large scale profiling of closely related substrates biasing the underlying data set towards non-specificity . A second bias of the selected set of investigated proteases is thereby inevitable: the selection of peptidases with more than 100 annotated cleavage sites in MEROPS favors well-studied as well as unspecific proteases . Hence , a putative perfectly specific protease cleaving only a single substrate and hence , cleavage entropy of zero , would not be covered in the presented test set . Proteases span a wide range of substrate specificites directly related to their biological roles . Ranking of the protease test set in respect to overall cleavage entropy SCleavage thus yields a clear separation between unspecific digestive proteases and specific proteases involved in signaling pathways . The protease with highest observed cleavage entropy SCleavage = 7 . 528 , thermolysin , is involved in bacterial nutrition by unspecificly degrading exogenous peptides [71] . The technical usage in protein sequencing [72] and peptide synthesis [73] is facilitated by this unspecific substrate recognition of thermolysin . On the other end of the test set's specificity spectrum , neurolysin is a primary example for a specific signaling protease with SCleavage = 4 . 477 . The limited proteolysis of intracellular oligopeptides by neurolysin [74] assures proper regulation of cell signaling [75] . An exemplary analysis of inter-subpocket correlation was carried out based on over 14000 trypsin substrates listed in MEROPS ( see Table S1 ) . Only pairs including the specific P1 position show pronounced imbalances in two-dimensional distributions of substrate amino acid pairs reflected in lowered pairwise cleavage entropy scores . All other subpocket pairs show pairwise cleavage entropies in the range of 0 . 896 to 0 . 923 implying low correlation between subpocket readout . If at all a cooperative effect can be detected between P1′ and P2 in the underlying dataset for trypsin ( SP1′ , P2 = 0 . 896 ) .
Strikingly , both extrema on the presented quantitative protease specificity scale for the core set of 47 proteases represent members of the metallo proteases ( thermolysin and neurolysin respectively ) . This indicates that the catalytic cleavage machinery cannot be the major determinant of substrate specificity . Similarly , serine proteases including the prominent digestive enzymes trypsin , chymotrypsin , elastase as well as signaling peptidases kexin and furin show diverse substrate specificity . Solely the smaller sample of five aspartic proteases shows predominantly unspecific cleavage behavior with an average total cleavage entropy of SCleavage = 7 . 205 compared to an average of SCleavage = 6 . 608 for the other catalytic types . Other protease classes do not show significant differences in their substrate specificity ( serine proteases: average SCleavage = 6 . 433 , metallo proteases: average SCleavage = 6 . 652 , cysteine proteases: average SCleavage = 6 . 820 ) . All protease types except for aspartic proteases therefore include specific as well as unspecific members . Thus , our study underlines the broadly accepted finding that protease substrate specificity is determined by subpocket interactions of the protease rather than directly at the catalytic site . As apparent from Figure 8 , the catalytic mechanism , does not discriminate specific from unspecific function . Rather , evolutionary related sub-groups sharing common catalytic mechanisms , but differing in three-dimensional fold are found to be similar in substrate promiscuity ( see Figure 9 ) . These clans within a catalytic class are not present in the test set for metallo proteases or aspartic proteases . All 13 metallo proteases in the test set belong to the MEROPS clan MA and all 5 aspartic proteases to the clan AA . Cysteine proteases spread over two distinct clans: 7 members ( cathepsins , calpains and falcipains ) belong to the CA papain clan , 3 others to clan CD , caspases . Serine proteases span three clusters of homologue proteases: 12 members are part of the PA clan ( chymotrypsin-like proteases ) , containing besides serine proteases also cysteine proteases , that are not covered within the test set . Two members of the clan SF share the signalase fold , whilst four others share a subtilisin fold and thus belong to MEROPS clan SB . Signal peptidase complex is not assigned to a particular MEROPS protease clan . Surprisingly , subdivision into homologue clans allows to subdivide proteases sharing the same catalytic mechanism into specific and unspecific subgroups . Cysteine proteases are divided into a more specific clan CD ( average SCleavage = 6 . 020 ) and a relatively unspecific clan CA ( average SCleavage = 7 . 163 ) . Only caspases , known to be highly specific signaling proteases [76] , represent clan CD in our test set , whereas calpains showing complex substrate specificities [41] with average SCleavage = 7 . 106 , cathepsins with average SCleavage = 7 . 113 or falcipains with SCleavage = 7 . 297 are contained in clan CA . Falcipains of malaria-causing Plasmodium falciparum are involved in cytoskeleton and hemoglobin degradation [77] requiring unspecific substrate binding . The same subdivision into specific and unspecific folds works for serine proteases that comprise clans of high specificity ( clan SB: average SCleavage = 5 . 429 ) , intermediate specificity ( clan SF: average SCleavage = 6 . 370 ) as well as less specific proteases ( clan PA: average SCleavage = 6 . 779 ) . Standard deviations of cleavage entropies calculated within clan members are low ( see Figure 9 ) , suggesting intrinsically encoded limits for specific/non-specific behavior within the three-dimensional fold of the respective clans . This finding could be attributed to an intrinsic presence or absence of preorganized subpockets allowing for specific enzyme-substrate interactions . Thus , the whole structure of protease clans has to be considered to shed light on the molecular origins of general protease cleavage spectra . Consistently , single mutations within specificity pockets of proteases are known to shift substrate spectra to other preferred substrates rather than to interchange specific and non-specific cleavage behavior . Nevertheless , a smooth interchange between specific and unspecific behavior including specialization and despecialization steps has been shown in case of granzymes [78] , a class of serine proteases in clan PA . Further tracing the evolutionary development of protease specificity into particular protease clans arises the question , if evolutionary distance at sequence level is related to substrate specificity in these groups with conserved three-dimensional fold . Therefore , we performed a phylogenetic analysis for individual protease clans with more than five members contained in the test set ( see Figure 10 ) . MEROPS protease families are grouped in branches , confirming reasonability of presented phylogenetic trees . Whereas all members of clan PA belong to family S1 , cysteine proteases spread over two distinct families: calpains are members of family C2 and are form a separate branch compared to all other proteases of the CA set that are part of the papain family C1 . Metallo proteases belong to a wide-spread range of families: neprilysin is a singleton of family M13 , neurolysin and thimet oligopeptidase of family M3 are nicely grouped in a separate branch . Two further singletons peptidyl-Lys metallo protease and thermolysin each form a separate tree branch for the families M35 and M4 respectively . All other members of clan MA are part of family M10 , the matrix metallo peptidases , and are grouped into a broad branch separated from the other members . Divergent evolution towards specific as well as unspecific members can be identified within all protease clans . Whereas a phylogenetic tree of metallo proteases of clan MA groups the highly specific members neurolysin and thimet oligopeptidase in a separate branch , indicating a close interplay between evolutionary distance and substrate specificity , this observation can not be extended to the whole set of proteases . The opposite holds even true in the MA clan for M10 family , where specific and unspecific members are grouped almost randomly compared to their evolutionary distance . The same complex behavior is found for cathepsins in clan CA: This branch includes the most specific member cathepsin L1 as well as the least specific member cathepsin K . Nevertheless , these members are grouped in closely related taxa indicating evolutionary proximity . Evolutionarily closely related proteases exhibit diverse substrate promiscuity in this protease group . Hence , protease evolution is capable of rapidly interchanging specific and non-specific substrate binding , implying a complicated relationship between protease sequence and substrate specificity . The largest group of serine proteases of clan PA also groups specific and unspecific members in related taxa . E . g . , cathepsin G and granzymes B of human and rodent origin exhibiting major different cleavage behavior are found as subbranch of closest evolutionary relation . Similarly , a branch including the rather specific signaling protease plasmin as well as the unspecific digestive enzymes trypsin 1 and chymotrypsin A , the most promiscuous members of this family , are grouped in close evolutionary proximity . We therefore surmise that a detailed understanding of protease specificity is only in reach within an even smaller subset of homologue proteases , where changes in substrate specificity can be attributed to a limited set of amino acid mutations , and hence atom exchanges , in the binding region . We propose to join forces between computational and experimental groups to elucidate structural hot-spots crucial for binding specificity in particular protease folds . According to the observed small fluctuations in specificity within respective clans , a smaller set of homologous proteases should be suitable to allow such in-depth investigations . The presented specificity metric “cleavage entropy” for proteases can be applied to map subpocket-wise specificity contributions based on experimental data to individual subpockets of proteases as well as to calculate an estimate of overall substrate specificity . Furthermore , the extension of subpocket-wise cleavage entropies to pairwise cleavage entropies facilitates the detection of subpocket cooperativities in proteases provided that a sufficient number of substrates for this two-dimensional analysis is known . Thereby , drug design targeting proteases will profit from a thorough understanding of specific interactions to achieve desired protease selectivity [79] for example in targeting matrix metallo proteases [80] . As parameters at the level of sequence [81] , structure [18] and conformational flexibility [82] are known to influence protease specificity , a direct quantification of substrate promiscuity of proteases will help to distinguish individual contributions to this phenomenon [83] and thereby support structural biology , the rational design of protease specificity [84] and the emerging field of degradomics [85] . An extension of the information-theory based specificity mapping towards general protein-protein interfaces to assess specificity and hence druggability of the respective interface regions is envisaged . A straight-forward interpretable specificity score generally applicable to all families of proteases was presented that confirms widely accepted rules of thumb for protease cleavage in a quantitative way . Calculated cleavage entropies purely based on amino acid frequencies in known substrates allow a straight-forward assessment of subpocket-wise substrate specificities . According to our specificity metric , the catalytic cleavage machinery and thus , protease class , does not discriminate specific and unspecific proteases . In contrast , homologue protease clans share intrinsic specific and non-specific properties suggesting that protease specificity is encoded directly in the shared three-dimensional protein fold . Within particular protease clans and folds , a small number of mutations can cause drastic alterations of substrate specificity . These subtle changes at sequence , structure and flexibility level , but heavily impacting substrate promiscuity , are thus of high interest for structural biology but challenging to predict . Unlike classical rules-of-thumb for protease specificity , the quantification of subpocket-wise and overall substrate specificity provides a continuous metric for specificity rather than a ‘yes’-or-‘no’ decision . The provided quantitative measure thus facilitates the comparison of the macromolecular descriptor “substrate specificity” with physicochemical , evolutionary and structural descriptors in protease recognition . Mapping of specificity to subpockets allows for intuitive visualization of structure-selectivity relationships in proteases and will thereby support the establishment of rules linking local protein structure and specificity . | Proteases show a broad range of cleavage specificities . Promiscuous proteases as digestive enzymes unspecifically degrade peptides , whereas highly specific proteases are involved in signaling cascades . As a quantitative index of substrate specificity was lacking , we introduce cleavage entropy as a measure of substrate specificity of proteases . This quantitative score allows for straight-forward rationalization of substrate recognition by a subpocket-wise assessment of substrate readout leading to specificity profiles of individual proteases as well as an estimate of overall substrate promiscuity . We present an exemplary application of the descriptor ‘cleavage entropy’ to trace substrate specificity through the evolution of different protease folds . Our score highlights the diversity of substrate specificity within evolutionary related proteases and hence the complex relationship between sequence , structure and substrate recognition . By taking into account the whole distribution of known substrates rather than simple substrate counting , cleavage entropy provides the unique opportunity to dissect the molecular origins of protease substrate specificity . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"biochemistry",
"protein",
"interactions",
"proteins",
"chemistry",
"biology",
"computational",
"chemistry"
] | 2013 | Cleavage Entropy as Quantitative Measure of Protease Specificity |
Eukaryotic protein kinases ( EPKs ) regulate numerous signaling processes by phosphorylating targeted substrates through the highly conserved catalytic domain . Our previous computational studies proposed a model stating that a properly assembled nonlinear motif termed the Regulatory ( R ) spine is essential for catalytic activity of EPKs . Here we define the required intramolecular interactions and biochemical properties of the R-spine and the newly identified “Shell” that surrounds the R-spine using site-directed mutagenesis and various in vitro phosphoryl transfer assays using cyclic AMP-dependent protein kinase as a representative of the entire kinome . Analysis of the 172 available Apo EPK structures in the protein data bank ( PDB ) revealed four unique structural conformations of the R-spine that correspond with catalytic inactivation of various EPKs . Elucidating the molecular entities required for the catalytic activation of EPKs and the identification of these inactive conformations opens new avenues for the design of efficient therapeutic EPK inhibitors .
Eukaryotic protein kinases ( EPKs ) phosphorylate a serine , threonine , or tyrosine residue in approximately 30% of human proteins and thus regulate numerous cellular and metabolic processes [1] . Abnormal catalytic activity of EPKs is implicated in numerous human diseases , including cancer , cardiovascular diseases , and diabetes . Therefore , EPKs are considered to be one of the most promising therapeutic drug targets . Of the more than 500 EPKs identified in the human genome , approximately 180 are associated with human diseases , either as causative agents or as therapeutic intervention points . Currently , 24 small molecule EPK inhibitors are FDA approved and numerous compounds are in clinical trials [2] . Some of the major challenges for designing efficient therapeutic drugs include the promiscuous nature of these inhibitors targeting multiple members of the family as well as patient relapse due to mutations that drive drug resistance [3] . EPKs have a highly conserved structural core that consists of two lobes: a small N-terminal lobe ( N-lobe ) and a larger C-terminal lobe ( C-lobe ) [4] , [5] . The smaller , N-lobe is primarily involved in anchoring and orienting the nucleotide ( Figure 1A ) . This lobe is predominantly constructed of antiparallel β-sheet structures that are unique among nucleotide binding proteins . A short loop known as the “hinge region” is the only structure that connects these two lobes . The deep cleft between the two lobes forms the active site where the phosphoryl transfer process occurs . Both the N- and C-lobes participate in the binding of ATP with 2 magnesium ions . The C-lobe binds the substrate , bringing it in close proximity to ATP , resulting in the phosphorylation of the substrate . Previous computational analysis of EPKs proposed that the core is organized around three major elements ( Figure 1B ) : a large hydrophobic αF-helix in the middle of the C-lobe and two nonlinear hydrophobic motifs termed “spines”: the Catalytic ( C ) spine and the Regulatory ( R ) spine [6] , [7] . The spines are anchored to the αF-helix and secure the position of ATP , substrate , and amino acid residues that are important for catalysis . The spines are unusual structural motifs as they consist of amino acid residues that come from different parts of the EPK sequence and do not form a conventional sequence motif . A unique feature of the C-spine is that the adenine ring of ATP is part of this spine and connects the hydrophobic residues from the N- and the C-lobes ( Figure 1B ) . The geometry of the R-spine is relatively stable as it remains intact throughout the phosphoryl transfer process . Unlike other enzymes , EPKs are unique as they do not have a single active and inactive conformation [8] . The active state of the enzyme is highly dynamic where the core toggles between the open and closed conformations . The inactive state has traditionally been divided into two general groups defined by the positioning of the phenylalanine of the DFG motif from the activation loop [9]–[11] . If the DFG-phenylalanine moves far enough from its active position , it is classified as the “DFG-out” conformation , which is currently the major target for therapeutic drug design . The second inactive conformation known as the “DFG-in” conformation is when the phenylalanine does not move substantially from the active conformation . The most common inactive DFG-in conformation is caused by the movement of the αC-helix , but other less understood inactive conformations , not caused by the movement of the DFG-motif or αC-helix , also belong to this group . Since the R-spine is a geometrically preserved motif that spans both lobes of all EPKs in the active state , we sought to elucidate the properties required for a catalytically functional R-spine ( Figure 1B ) . Using an E . coli expression system , site-directed mutagenesis , Western blotting , and a radioactive phosphoryl transfer assay , we elucidated the biochemical and biophysical properties required for a catalytically viable R-spine using cyclic AMP-dependent protein kinase ( PKA ) as a model system . We identified three additional hydrophobic residues around the R-spine that we refer to as the “Shell , ” which play a crucial role in supporting the R-spine's ability to maintain catalytic function . We experimentally tested the relationship between the phosphorylation state of the activation loop , the R-spine , and the catalytic activity in PKA . Additionally , qualitative structural analysis of 172 available Apo EPK structures from the protein data bank ( PDB ) lead to the identification of four distinct ways the R-spine is disassembled corresponding with catalytic inactivation of EPKs .
In most EPKs the R-spine consists of two aromatic residues from the C-lobe and two aliphatic residues from the N-lobe . Alignment of more than 13 , 000 EPK sequences ( Table S1 ) showed that RS1 is conserved as an aromatic residue in ∼99% of EPKs ( Table 1 ) . Previous studies in Drosophila Src64 showed that mutating RS1 from a histidine into a leucine does not eliminate the catalytic activity [17] . To examine whether PKA can tolerate an aliphatic residue instead of a tyrosine at the RS1 position , we inserted a mutation converting RS1 to a methionine ( RS1M ) . The Western blot assay illustrates that the mutant remains catalytically active even though activity is reduced ( Figure 2A , G ) . The second R-spine residue from the C-lobe , RS2 , is conserved as a phenylalanine in approximately ∼90% of EPKs , but this residue is also a leucine ( aliphatic ) in about ∼6 . 7% of EPKs . To test if this naturally occurring variant is tolerable in PKA , we mutated RS2 to a leucine ( RS2L ) . We observed that RS2L remains as catalytically active as the wild-type PKA ( WT-PKA ) ( Figure 2A , G ) . The N-lobe residues RS3 and RS4 , on the other hand , are conserved as aliphatic residues in approximately 90% and 80% of EPKs , respectively . In order to determine if PKA can tolerate an aromatic residue instead of an aliphatic residue , we individually mutated RS3 and RS4 from a leucine into a phenylalanine ( RS3F and RS4F ) . The results show that RS3F and RS4F mutants have normal levels of catalytic activity in comparison to the WT-PKA ( Figure 2A , G ) . The RS1 residue is conserved as a histidine in ∼91% and tyrosine in ∼8% of EPKs ( Table 1 ) . The side chain of RS1 has the ability to interact with the neighboring amino acid residues in three different ways: first , the hydrophobic interaction of tyrosine with RS2 , which we demonstrated to be sufficient for maintaining some catalytic activity ( RS1M ) ( Figure 2B , G ) . Next is the CH-π interaction with RS2 , which is conserved in approximately 90% of EPKs . To assess whether this interaction is sufficient for maintaining catalytic activity , we replaced RS1 with phenylalanine ( RS1F ) . The Western blot results show that RS1F is sufficient for maintaining catalytic activity , and the introduction of the aromatic ring improves the catalytic activity when compared to RS1M ( Figure 2B , G ) . Since RS1 is conserved as histidine ( RS1H ) in 91% of EPKs , we mutated RS1 from a tyrosine to histidine to see if these two residues were interchangeable , and Western blot analysis shows that the mutant remains catalytically active ( Figure S1 ) . Finally , there is the polar interaction of RS1 with RS0; here , the main chain of RS1 interacts with the side chain of RS0 and this interaction is conserved in more than 95% of EPKs . To examine if this polar interaction is required to maintain the proper assembly of the R-spine , we mutated RS0 to alanine ( RS0A ) , and Western blot results show that catalytic activity was abolished ( Figure 2B , G ) . This is consistent with recent studies on Aurora kinase , where mutation of RS0 to an alanine ( RS0A ) abolished Aurora kinase activity [18] . To test if loss of this polar interaction could be rescued through a hydrophobic interaction , we replaced RS0 with a leucine ( RS0L ) ; results show that some catalytic activity was rescued ( Figure 2B , G ) . Based on sequence alignment ( Table 1 ) , the three R-spine residues ( RS2 , RS3 , and RS4 ) are highly conserved as a hydrophobic residue , whereas only ∼8% of EPKs including PKA have a hydrophobic residue at the RS1 position . To address if the hydrophobic property is required for catalytic activity , we introduced the hydrophilic residues aspartic acid ( RS1D , RS2D , RS3D , and RS4D ) or asparagine ( RS1N , RS2N , RS3N , and RS4N ) to each of the four R-spine positions individually . Using Western blotting techniques and radioactive phosphoryl transfer assays , we discovered that the two C-lobe residues ( RS1 and RS2 ) were highly sensitive to the introduction of a hydrophilic residue . Western blot analysis demonstrates that the catalytic activity was abolished when RS1 or RS2 were substituted with a hydrophilic residue ( RS1D , RS2D , RS1N , and RS2N ) ( Figure 2C , G ) . Quantitative analysis of the hydrophilic noncharged asparagine mutation using the radioactive phosphoryl transfer assay confirmed that the catalytic activity of RS1N and RS2N was reduced by more than 95% ( Figure 2D , G and Table S2 ) . In contrast , when the N-lobe R-spine residues were mutated to hydrophilic residues ( RS3D , RS4D , RS3N , and RS4N ) , the mutants remained catalytically active ( Figure 2C , G ) . The enzyme retained 85% and 95% of its activity when the RS3 and RS4 positions were mutated to asparagine ( RS3N and RS4N ) , respectively ( Figure 2D , G and Table S2 ) . To evaluate whether the side chain of each R-spine residue is required for catalytic activity , we individually mutated each residue into an alanine or a glycine . After mutating the RS1 and RS2 ( RS1A , RS2A RS1G , and RS2G ) , the catalytic activity was abolished , as illustrated by the Western blots ( Figure 2E , G ) . The radioactive phophoryl transfer assay confirmed that the catalytic activity was reduced by more than 99% for RS1G and RS2G ( Figure 2F , G and Table S2 ) . However , the Western blots of RS3 and RS4 to alanine and glycine mutants ( RS3A RS4A , RS3G , and RS4G ) showed that these mutants had comparable levels of catalytic activity as the WT-PKA ( Figure 2E , G ) . The quantitative data for RS3G and RS4G confirm that the catalytic activity was only reduced by ∼15% and ∼5% , respectively ( Figure 2F , G and Table S2 ) . To understand why the catalytic activity was unaffected by the introduction of a hydrophilic residue or the removal of the side chain to the N-lobe region of the R-spine , we analyzed the amino acid residues that are within 4 Å of RS3 and RS4 in PKA . Looking at the previous Local Spatial Pattern ( LSP ) alignment data [7] , only three out of the 14 amino acid residues surrounding RS3 and RS4 are highly conserved . We termed these three residues as the Shell , as they seemed to be supporting the N-lobe region of the R-spine ( Figure 3A ) . In PKA these residues are valine 104 ( Sh1 ) , which is conserved as a hydrophobic residue in ∼90% of EPKs , the gatekeeper residue ( methionine 120 ( Sh2 ) ) , which is conserved as a hydrophobic residue in ∼82% of EPKs , and methionine 118 ( Sh3 ) , conserved as a hydrophobic residue in ∼98% of EPKs ( Table 1 and Figure 3B ) . To understand the role of the Shell for catalytic activity , we made multiple mutations followed by radioactive phosphoryl transfer assays in PKA . Above , we showed that RS3G has comparable catalytic activity to the WT-PKA ( Figure 3C , D and Table S2 ) . To destabilize this mutant , we decided to introduce an alanine mutation at the Sh2 position ( RS3G/Sh2A ) . Results showed that the catalytic activity was significantly reduced by ∼96% ( Figure 3C , D and Table S2 ) . This indicates that RS3 is essential in the absence of Sh2 . Next , to understand the role of Sh1 on the catalytic activity , we mutated Sh1 into a glycine ( Sh1G ) . In the absence of the Sh1 side chain , the catalytic activity of the EPK was reduced by ∼94% ( Figure 3C , D and Table S2 ) . Sh1 is a crucial residue for catalytic activity as previous studies describe that the αC-β4-loop is crucial for anchoring the αC-helix [19] . To understand the significance of each of the three residues ( RS4 , Sh3 , and Sh2 ) , we mutated the RS4 residue to a glycine and SH2 and SH3 residues to an alanine ( RS4G/Sh2A/Sh3A ) . This catalytically dead triple mutant serves as the reference point for understanding the role of each residue on the catalytic activity of the EPK ( Figure 3C , D and Table S2 ) . We then reintroduced each residue individually into the triple mutant to assay for rescue of catalytic activity . When returning Sh2 to a methionine in the triple mutant ( RS4G/Sh3A ) , we were able to rescue ∼13% of the catalytic activity ( Figure 3C , D and Table S2 ) . Next , we mutated RS3 to a glycine ( RS4G/Sh3A/RS3G ) to address whether Sh2 could maintain activity through Sh1 . Here we were able to rescue the catalytic activity by ∼23% , which indicates that Sh1 plays a role in maintaining a viable hydrophobic interaction between the N- and C-lobes through Sh2 ( Figure 3C , D and Table S2 ) . Next , we attempted to rescue some of the catalytic activity by mutating Sh3 back to a methionine in the triple mutant ( RS4G/Sh2A ) , and we were able to rescue ∼47% of the catalytic activity ( Figure 3C , D and Table S2 ) . Above , we showed that in the absence of Sh2 and RS3 , the catalytic activity was abolished , and this indicates that Sh3 requires the presence of RS3 to maintain catalytic activity ( Figure 3C , D and Table S2 ) . Finally , when we returned RS4 back to a leucine in the triple mutant ( Sh3A/Sh2A ) , we were unable to recover any catalytic activity ( Figure 3C , D and Table S2 ) . This indicates that RS4 requires the presence of either Sh2 or Sh3 to maintain catalytic activity . Complete activation of PKA is achieved after phosphorylation of T197 ( pT197 ) on the activation loop [20] . pT197 initiates a major hydrogen bonding network in the C-lobe [13] and forms a H-bond between the activation loop and the αC-helix through H87 . Previous studies showed that eliminating the H-bond between pT197 and H87 improves the catalytic activity by 2–3-fold [14] . We hypothesized that destabilization of the R-spine through hydrophilic mutations would cause disorientation of the N- and C-lobes , and this loss of catalytic activity can be rescued through the H-bond formed between pT197 and H87 ( Figure 4A , B ) . To test this hypothesis we began by rescuing the catalytic activity of RS1N and RS2N by co-expressing these constructs with PDK1 , which phosphorylates PKA on T197 and introduces the pT197-H87 H-bond . Using a radioactive phosphoryl transfer assay , we observed a ∼43% rescue of catalytic activity from ∼2% for RS1N and ∼75% rescue from less than 1% catalytic activity for RS2N ( Figure 4C and Table S2 ) . As a control we co-expressed RS3N with PDK1 ( RS3+PDK1 ) , and results show that the catalytic activity is comparable with RS3N . To understand if this rescue of function was due the intricate hydrogen bond network formed in the C-lobe or due to pT197-H87 , we introduced the H87A mutation into all the asparagine mutants and co-expressed these double mutants with PDK1 ( RS1N/H87A+PDK1 , RS2N/H87A+PDK1 , RS3N/H87A+PDK1 , and RS4N/H87A+PDK1 ) . The results from the radioactive phosphoryl transfer assay show that the catalytic activity for RS2N/H87A+PDK1 was reduced by ∼95% in comparison to RS2N+PDK1 and the catalytic activity for RS3N/H87A+PDK1 was reduced by ∼73% with respect to RS3N+PDK1 ( Figure 4C and Table S2 ) . However , the catalytic activity of RS1N/H87A+PDK1 and RS4N/H87A+PDK1 was reduced by ∼50% and ∼55% , when compared to RS1N+PDK1 and RS4N , respectively ( Figure 4C and Table S2 ) . Although the effect is not as drastic for RS1 and RS4 , these results demonstrate that any instability of the R-spine affects the catalytic activity .
In 2006 , the R-spine hypothesis for EPK regulation was proposed based on the computational comparison of 23 EPK structures [6] . Despite the lack of solid biochemical validation of this model , it quickly became popular and has been widely accepted as a framework for analysis of EPKs [21]–[24] . Nevertheless , many questions related to the properties of the R-spine residues remained unanswered . In this work we present the first systematic study of the R-spine in PKA , an EPK that has served as a prototype for the entire kinome for more than two decades . Our findings establish that the hydrophobic nature of the R-spine and the nonpolar CH-π interaction of RS1 with RS2 are mandatory for catalytic activity . The interaction of RS0 with RS1 is crucial for catalytic activity , but we demonstrate that a hydrophobic interaction can maintain the anchoring of the R-spine to the αF-helix . We also revealed that the N-lobe region of the R-spine is supported by a three-residue hydrophobic ensemble that we termed the “Shell” ( Sh1 , Sh2 , and Sh3 ) . The absence of Sh1 causes catalytic inactivation , indicating that the interaction of Sh1 with RS3 is crucial for anchoring the αC-helix in the active conformation . Sh1 is also capable of maintaining some catalytic activity in the absence of RS3 by completing the R-spine through the gatekeeper residue ( Sh2 ) . Previous studies showed that Sh2 and Sh3 are equally important for catalytic activity because either residue has the ability to compensate for the absence of the other , as in IL2-inducible T-cell kinase ( Itk ) , and that at least one was mandatory for maintaining catalytic activity [25] . Here we confirm that the absence of Sh3 and Sh2 in PKA abolishes catalytic activity and returning either one enables the partial rescue of catalytic function . This is supported by the numerous disease-driving bulky hydrophobic single-nucleotide polymorphisms of the gatekeeper ( Sh2 ) residue that boost activity [21] . The absence of a perfectly assembled R-spine results in loss of catalytic activity . However , this loss of function can be rescued by phosphorylating the activation loop , thus creating the pT197-H87 H-bond that stabilizes the assembled conformation of the R-spine . Since the assembly of the R-spine is required for catalytic activity , we searched for naturally occurring disassembled conformations of the R-spine that correlated with catalytic inactivation . From the available 172 Apo EPK structures available in the PDB , we identified four different ways the R-spine can be disassembled corresponding to catalytically inactive EPKs ( Table S3 ) . The first two inactive groups were described as the DFG-out and DFG-in inactive conformations , respectively [9] . Inactive I or the DFG-out conformation is where the side chain of the RS2 of the DFG motif is misplaced from the active conformation as illustrated by the structure of Protein kinase B ( AKT ) [26] ( Figure 5 , Table S3 ) . This inactive conformation was previously described in ABL kinase , and we were able to mimic this conformation in PKA through the RS2G mutant . Inactive II or the αC-helix out conformation , which was previously described in Src , occurs when RS3 is removed from the active conformation due to the αC-helix twisting out and moving away from the active site ( Figure 5 and Tables S3 and S4 ) The RS3G+Sh2A mutant mimics this inactive conformation as the Sh2 residue in Src is the small hydrophilic residue threonine . In the inactive III or the YRD/HRD-out conformation , represented by 5′ AMP-activated protein kinase [27] , we observe that RS1 from the YRD/HRD motif of the catalytic loop is no longer anchored to the αF-helix ( Figure 5 , Tables S3 and S4 ) . We were able to mimic this conformation through the RS0A mutant in PKA . The last inactive conformation is the inactive IV or the twisted lobe conformation , which occurs when the two lobes move away from each other and twist , causing the R-spine to split in half ( Figure 5 , Tables S3 and S4 ) . This conformation is represented by the structure of P38 mitogen-activated protein kinases [28] . Identification of the inactive I and inactive II conformations enabled the design of successful drugs such as Imatinib [29] and Lapatinib [30] , respectively ( Figure S2 ) . We believe that the identification and the functional understanding of the R-spine and Shell will generate novel approaches to designing more efficient therapeutic EPK inhibitors as well providing insight towards understanding some of the disease-causing mutations .
Representative EPK sequences from major taxonomic groups and families ( ∼13 , 690 sequences ) were identified and multiply aligned using the MAPGAPS program [31] . The aligned columns were used to calculate amino acid frequencies at each of the R-spine positions ( RS0–RS4 ) and the Shell positions ( Sh1–3 ) . QuikChange II site-directed mutagenesis kit ( Agilent technologies ) was used to introduce various mutations . The His6-tagged murine Cα-subunit of cAMP-dependent protein kinase ( PKA ) in pET15b was expressed in E . coli ( BL21 ( DE3 ) ) . Cultures were grown at 37°C to an A600 of ∼0 . 6 and induced with 0 . 5 mM isopropyl β-d-thiogalactopyranoside ( IPTG ) . The cultures were allowed to grow overnight at 16°C before harvesting . The expression of PKA was confirmed using PKA C-subunit antibodies from BD Biosciences , and the phosphorylation state of the activation loop was confirmed using a polyclonal pT197 antibody from Invitrogen . The His6-tagged wild-type and mutation containing PKA in pET15b as well as mutants co-expressed with GST-tagged PDK1 were expressed in E . coli ( BL21 ( DE3 ) ) . Cultures were grown at 37°C to an A600 of ∼0 . 6 and induced with 0 . 5 mM IPTG . The cultures were allowed to grow overnight at 16°C before being harvested . The pellet was resuspended in lysis buffer ( 50 mM KH2PO4 , 20 mM Tris-HCl , 100 mM NaCl , 5 mM β-mercaptoethanol , pH 8 . 0 ) and lysed using a microfluidizer ( Microfluidics ) at 18 , 000 p . s . i . The cells were clarified by centrifugation at 15 , 000 rpm at 4°C for 60 min in a Beckman JA20 rotor , and the supernatant was incubated with TALON His-Tag Purification Resin ( Clontech ) overnight at 4°C using gravity . The resin was washed twice ( 20× bed volume ) with the lysis buffer and twice with using two different concentrations of imidazole in the wash buffer ( 50 mM KH2PO4 , 20 mM Tris-HCl , 100 mM or 1 m NaCl , 50 mM/100 mM imidazole , and 5 mM β-mercaptoethanol , pH 7 ) . A 250 mM imidazole elution buffer was used to elute the His-tagged protein ( Figure S3A ) . The kinetics was carried out with common reaction mix containing 50 mM MOPS pH 7 . 4 , 1 mM Kemptide , 10 mM MgCl2 , 1 mM ATP , and 32 γP radiolabelled ATP ( specific activity 500–1 , 000 cpm/pmol ) in a final volume of 20 µL . The reaction was initiated by addition of PKA with final concentration of 50 nM in a volume of 10 µl to 10 µl of the reaction mix described above . The reaction was carried out as an end point assay with 3 min as fixed time , and at the end-point the reaction was quenched with 90 µl of 30% Acetic acid . 50 µl of the quenched reaction was then spotted on p81 phosphocellulose paper , washed three times for 5 min each with 5% phosphoric acid , and finally washed with acetone ( 1× ) ; air dried; and counted on liquid scintillation counter . The background counts were subtracted from the experimental time points and plotted to compare their activities . The wild-type protein was used as positive control and E91A mutant as the negative control . Each reaction was carried out in triplicates and data plotted are mean percent of WT-PKA ± relative standard error . The plots were made using MS Excel . The His6-tagged murine Cα-subunit of PKA containing catalytically inactive mutants in pET15b was co-expressed with GST-tagged PDK1 in E . coli ( BL21 ( DE3 ) ) . PDK1 required properly folded PKA to phosphorylate PKA on the activation loop T197 [32] . Cultures were grown at 37°C to an A600 of ∼0 . 6 and induced with 0 . 5 mM IPTG . The cultures were allowed to grow overnight at 16°C before harvesting . The expression of PKA was confirmed using PKA C-subunit antibodies from BD Biosciences , and the phosphorylation state of the activation loop was confirmed using a polyclonal pT197 antibody from Invitrogen ( Figure S3B ) . | Eukaryotic protein kinases ( EPKs ) have a highly conserved enzymatic kinase core that is involved in the regulation of numerous cell signaling processes through the transfer of a phosphate group from adenosine triphosphate ( ATP ) to more than 30% of human proteins . EPKs have been implicated in numerous human diseases , including cancer , cardiovascular diseases , and diabetes , making them one of the most sought-after therapeutic drug targets . The lack of structural diversity of the active kinase core has created a bottle-neck for designing successful therapeutic inhibitors . Here we describe the intramolecular interactions required for differentiating between the active and inactive states of EPKs . Kinases contain a hydrophobic regulatory spine ( “R-spine” ) that is disassembled in inactive kinases , and here we define an additional hydrophobic “Shell” that surrounds one end of the R-spine . Biochemical analysis of the five nonconsecutive R-spine residues and three nonconsecutive Shell residues shows that proper assembly of the R-spine and Shell is essential for maintaining kinase activity . Structural analysis of the 172 known structures of EPKs without bound ligands led to the identification of four inactive conformations that correlate with the disassembly of the R-spine . Understanding the molecular elements involved in the regulation of kinase activity and the identification of these diverse groups of inactive conformations should aid the design of more specific therapeutic EPK inhibitors . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Deciphering the Structural Basis of Eukaryotic Protein Kinase Regulation |
New foci of human CL caused by strains of the Leishmania donovani ( L . donovani ) complex have been recently described in Cyprus and the Çukurova region in Turkey ( L . infantum ) situated 150 km north of Cyprus . Cypriot strains were typed by Multilocus Enzyme Electrophoresis ( MLEE ) using the Montpellier ( MON ) system as L . donovani zymodeme MON-37 . However , multilocus microsatellite typing ( MLMT ) has shown that this zymodeme is paraphyletic; composed of distantly related genetic subgroups of different geographical origin . Consequently the origin of the Cypriot strains remained enigmatic . The Cypriot strains were compared with a set of Turkish isolates obtained from a CL patient and sand fly vectors in south-east Turkey ( Çukurova region; CUK strains ) and from a VL patient in the south-west ( Kuşadasi; EP59 strain ) . These Turkish strains were initially analyzed using the K26-PCR assay that discriminates MON-1 strains by their amplicon size . In line with previous DNA-based data , the strains were inferred to the L . donovani complex and characterized as non MON-1 . For these strains MLEE typing revealed two novel zymodemes; L . donovani MON-309 ( CUK strains ) and MON-308 ( EP59 ) . A population genetic analysis of the Turkish isolates was performed using 14 hyper-variable microsatellite loci . The genotypic profiles of 68 previously analyzed L . donovani complex strains from major endemic regions were included for comparison . Population structures were inferred by combination of Bayesian model-based and distance-based approaches . MLMT placed the Turkish and Cypriot strains in a subclade of a newly discovered , genetically distinct L . infantum monophyletic group , suggesting that the Cypriot strains may originate from Turkey . The discovery of a genetically distinct L . infantum monophyletic group in the south-eastern Mediterranean stresses the importance of species genetic characterization towards better understanding , monitoring and controlling the spread of leishmaniasis in this region .
Leishmaniases are a group of neglected infectious diseases caused by obligate intracellular protozoa of the genus Leishmania and transmitted by sand flies of the Phlebotominae subfamily . They are characterized by a spectrum of clinical manifestations ranging from ulcerative skin lesions ( cutaneous leishmaniasis , CL ) to a life-threatening disseminated visceral infection ( visceral leishmaniasis , VL ) . The overall prevalence of leishmaniasis is estimated to 12 million cases worldwide , and the global yearly incidence of all clinical forms is two million cases [1] . In the Old World , Leishmania major and L . tropica are the prevalent dermotropic species causing CL , whereas strains belonging to the L . donovani complex are typically responsible for VL . The current nomenclature of this complex encompasses only two species , L . donovani and L . infantum . L . archibaldi previously described for strains isolated in East African VL foci has been shown to be an invalid species [2] , and L . chagasi was found to be virtually identical to L . infantum from Southern Europe and recently introduced to the Americas [3] . Some L . infantum variants found in Mediterranean countries are predominantly dermotropic [4] , [5] , [6] . In the Eastern Mediterranean Region ( EMR ) leishmaniasis represents a major public health problem with considerable impact on morbidity and the potential to spread . Zoonotic and anthroponotic CL caused by L . major and L . tropica , respectively account for the largest number of cases in this region , although zoonotic VL caused by L . infantum is also common . Nevertheless , data on causative agents , vectors or reservoirs are not regularly available [7] . In this context , Turkey a country that lies at the crossroad between Asia and Europe represents a geographic site of special epidemiological interest regarding leishmaniasis . Both zoonotic VL caused by L . infantum and anthroponotic CL due to L . tropica have long been known to exist in several regions of Turkey [8] . VL is endemic throughout the Aegean , Central Anatolia , Marmara , Mediterranean and Black Sea Regions and CL is hyperendemic in south-eastern parts of Turkey neighbouring Iraq and Syria . According to Multilocus Enzyme Electrophoresis ( MLEE ) , which remains the reference method for typing Leishmania strains [1] , L . infantum MON-1 is the prevalent zymodeme causing VL in Turkey , as in all Mediterranean countries . CL cases caused by L . infantum spp . are mainly reported from the Çukurova region of south-east Turkey . In particular , Multilocus Sequence Typing ( MLST ) followed by phylogenetic analysis assigned a single strain isolated from a CL human case and two Phlebotomus tobbi ( P . tobbi ) sand fly isolates from the Çukurova region in Adana province ( CUK1 , CUK2 and CUK10 ) to L . infantum . However , the isolates did not group with the MON-1 clade but with a MON-188 strain ( ISS800 ) from Italy ( Sicily ) [6] . CL cases due to species of the L . donovani complex have also been reported in adjacent areas of Middle East countries , such as Lebanon and Syria , where the causative agents were identified as L . infantum by MLEE typing [9] . In another study a CL isolate from Lebanon was characterized as L . archibaldi [10] . Also , it was shown that both L . infantum and L . donovani parasites circulate in the northwest of Iran [11] . Notably , CL cases due to L . donovani MON-37 were also detected in Cyprus [12] , [13] . Interestingly , most reported CL cases including those from Turkey fall within a 500 km radius area . All have been typed at species level mostly by PCR-based molecular methods , which for practical reasons have replaced MLEE . It is therefore possible that in the past , due to the lack of easily applicable Leishmania PCR-typing tools , some CL cases identified in this region were falsely attributed to the traditional dermotropic species ( L . major , L . tropica ) based solely on clinical criteria . Hence , strains of the L . donovani complex could potentially have a substantial contribution to the generation of CL in the EMR . All the above point to the need for an in-depth study of the genetic diversity of parasite populations in the region , at different levels ( genus , complex , species , population or even strain ) . This can be achieved using methods with high discriminatory power and reproducibility that allow inter-lab comparisons [14] , [15] such as multilocus microsatellite typing ( MLMT ) . Notably , MLMT has been extensively used for population genetic studies in Leishmania throughout the world giving useful insights into the epidemiology of leishmaniasis [3] , [16] , [17] , [18] . This approach can contribute to a better understanding of the geographical distribution and dynamics of Leishmania populations and disease epidemiology in the EMR . In this context , we have analyzed a set of L . donovani complex strains isolated from human CL foci in Çukurova and from a VL patient from Kuşadasi in the Aegean region of Turkey . This set was compared to strains from Cyprus causing VL or CL that were MLEE- typed as L . donovani MON-37 [12] but shown by MLMT to be genetically distinct from MON-37 strains of other regions [16] . The strains were first analyzed using the K26-assay that discriminates L . donovani complex subspecies and then typed further using MLEE . In order to elucidate their population structure , the MLMT approach was applied . The microsatellite profiles of 68 previously analyzed Leishmania donovani complex strains from major endemic regions of VL , including those from Cyprus , were used in the analysis for comparison .
Table 1 lists the 76 L . infantum and L . donovani strains used in this study along with their WHO code , geographic origin , zymodeme type , pathology and population assignment . The Leishmania strains from Turkey included EP59 [19] isolated from a VL patient in Kuşadasi province in the Aegean region of Turkey ( nearby Izmir ) and six strains from Çukurova region in south-eastern Turkey [6] . The latter were five isolates from P . tobbi ( CUK2 , CUK3 , CUK4 , CUK7 , CUK10 ) and a single isolate from a human CL case ( CUK1 ) . We have also included three MON-37 clones of CD44 , isolated in 2005 from a dog in Cyprus . Strains EP59 , CUK1 , CUK2 and CUK3 were typed by MLEE using the Montpellier method ( MON ) , as described by Rioux et al . [20] . The microsatellite profiles and genetic groups of 68 previously analyzed strains belonging to the major L . donovani populations , the non MON-1 L . infantum population and the MON-1 population have been described in previous publications [16] , [17] , [18] , [21] . These L . infantum and L . donovani strains were included for comparison because they originate from various geographic regions and reflect the zymodeme diversity of the two species . The strains were grown as promastigotes at 26°C in RPMI 1640 containing 20 µmol/L HEPES buffer ( GIBCO-BRL Paisley , UK ) , supplemented with 2 mmol/L glutamine , 10% heat-inactivated fetal bovine serum , 100 IU/mL penicillin , and 100 µg/mL streptomycin . Total parasite DNA was extracted from mass cultures or clinical samples using phenol/chloroform [22] , suspended in water , and stored at −20°C . DNA was isolated from stocks that in their majority were subjected to limited in vitro passages . Based on previous preliminary results [12] we decided to clone , strain CD44 isolated in 2005 from a dog in Cyprus ( unpublished data , Gouzelou et al . ) , and EP59 isolated in 2001 from a VL patient in south-west Turkey . Cloning of Leishmania promastigotes from cultures was carried out by the hanging drop method [23] . In brief , single parasites ( examined under the microscope independently by two researchers for confirmation ) were isolated from minute drops of serially diluted parasite cultures , grown in 96-well plates in RPMI and subsequently on Novy-MacNeal-Nicolle ( NNN ) [23] , [24] . The K26-PCR assay is specific to the L . donovani complex and is capable of discriminating L . infantum MON-1 from other L . donovani complex subspecies , based on length polymorphism of K26 gene . This assay was applied on clones of CD44 , the CUK isolates as well as on parent and clones of the EP59 strains . Amplification reactions and subsequent determination of the amplicon size were carried out as described previously [25] . Fourteen variable microsatellite markers randomly distributed throughout the genome ( Li 22-35 , Li 23-41 , Li 41-56 , Li 45-24 , Li 46-67 , Li 71-5/2 , Li 71-7 , Li 71-33 , Lm2TG , Lm4TA , TubCA , CS20 , kLIST 7031 , kLIST 7039 ) were used in the present study as previously described [21] . For the amplification of a microsatellite locus , one of each pair of the HPLC-purified primers ( Proligo , France ) used , was conjugated at its 5′ end to one of three fluorescent dyes . PCR reactions and amplifications were performed using the PTC-200 thermocycler ( MJ Research Inc . , Watertown , MA ) as described elsewhere [21] , [26] . The amplified fragments were subjected to automated fragment analysis on the capillary sequencer ( CEQ 8000; Beckman Coulter ) and analysed with the AFLP analysis software . To infer the population structure of the sample set , the multilocus microsatellite genotypes were analysed using a model-based clustering method implemented in STRUCTURE v 2 . 3 . 1 [27] . This algorithm can assign individuals to populations probabilitistically , based on their multilocus genotypes , and can estimate the posterior probability for a given number of genetic populations ( K ) thus enabling the identification of the most likely number of populations . The admixture model with correlated allele frequencies was assumed and a burn-in time of 20 , 000 followed by 200 , 000 iterations was used . Ten independent runs for each K were carried out for each possible number of clusters ( K ) in order to quantify the variation in the likelihood of the data for a given K . The most probable number of genetic clusters was estimated by comparing log-likelihood values for K between 1 and 12 . The major population structure was captured at the plateau ( maximum ) of the derived Gaussian graph . In addition , as suggested by Evanno et al . to better determine the number of populations , ΔK was calculated for each K [28] , which is based on the second-order rate of change in the log probability of data with respect to successive K values . Microsatellite-based genetic distances were calculated using the software Microsat [29] by applying the proportion of shared alleles ( Dps ) distance measure [30] . Dps is defined as the negative logarithm of the proportion of shared alleles and follows the infinite allele model ( IAM ) . On the basis of the calculated distance matrices , a midpoint-rooted neighbour-joining ( NJ ) tree was constructed using PHYLIP , v 3 . 6 [31] . Confidence intervals were obtained by bootstrapping ( 1000 replicates ) . For the construction of a consensus tree and additional tree editing the programs Geneious [32] and Figtree ( http://tree . bio . ed . ac . uk/software/figtree ) were utilized . To further characterise the genetic substructure at both population and individual level , a factorial correspondence analysis ( FCA ) implemented in GENETIX v 4 . 03 software [33] was carried out . This analysis places the individuals in three-dimensional space according to the degree of their allelic state similarities . In order to analyse the populations defined by STRUCTURE with respect to diversity of alleles ( A ) , expected heterozygosity ( He ) , observed heterozygosity ( Ho ) and inbreeding coefficient FIS , the GDA software package was applied ( http://hydrodictyon . eeb . uconn . edu/people/plewis/software . php ) . Strain collection and Leishmania DNA used in this study , as well as the respective ethical approvals , were described in previous publications [12] , [16] , [17] , [18] , [19] , [21] . All samples were anonymized . The present study was approved by the Ethics Committee of the Hellenic Pasteur Institute .
The K26-PCR assay was the method of choice for initially characterizing the Turkish human and sand fly isolates from the CL foci in Çukurova as well as the VL isolate from Kuşadasi . A K26 amplicon of 385 bp was obtained for the EP59 strain and its clones . The same size for the K26 amplicon has been previously observed in the MHOM/MT/85/BUCK strain [25] , an L . infantum MON-78 strain from Malta that was also included in our analysis for comparison ( Table 1 , Fig . 1 ) . All 6 strains from the Çukurova region presented a 480 bp K26 amplicon , which was observed for the first time . The isolates from Cyprus including the MON-37 strain ( CH35 ) used as reference herein gave a 700 bp amplicon [12] . A 626 bp K26 amplicon typical for strains of zymodeme MON-1 was yielded for the PMI strain from Spain as expected ( Fig . 1 ) . Subsequent MLEE analysis of four of the Turkish isolates revealed two novel L . donovani zymodemes . The EP59 strain was typed as MON-308 with enzymatic patterns identical to those of L . infantum MON-1 except for the two glutamate-oxaloacetate transaminases ( GOT1 and GOT2 , EC 2 . 6 . 1 . 1 ) . In particular , iso-electrofocusing showed a heterozygous pattern between the L . infantum GOT100 and the L . donovani GOT113 ( F . Pratlong , unpublished data ) . The three isolates from Çukurova , namely CUK1 , CUK2 and CUK3 presented exactly the same enzymatic patterns and were typed as L . donovani MON-309 . This variant had identical isoenzyme electrophoretic mobilities with the L . donovani MON-3 zymodeme apart from malate dehydrogenase ( MDH , EC 1 . 1 . 1 . 37 ) that presented a relative mobility of 145 . This is distinct from all other known electrophoretic variants of MDH found among L . donovani complex strains . Compared to the MON-37 zymodeme , the novel MON-309 zymodeme differs in the mobilities of just two isoenzymes , notably MDH and glucose phosphate isomerase ( GPI , EC 5 . 3 . 1 . 9 ) . Using a set of 14 high-resolution microsatellite L . donovani complex specific markers [26] all newly analyzed Turkish strains ( TR ) presented unique genotypes and significant intra-zymodeme diversity was detected . Among strains of the newly identified MON-309 zymodeme different levels of allelic variation were observed that range from a single difference in 1 allele ( strains CUK1 and CUK2 ) to variations in up to 8 alleles ( strains CUK2 and CUK10 ) . Interestingly the EP59 strain was characterised by an unusual high number of heterozygous loci ( 9 out of 14 loci studied ) . In each of these loci one allele was typical MON-1 and the second corresponded to the allele sizes found for the CUK strains ( Table S1 ) . In order to infer the population structure/substructure of our sample set the MLMT profiles of the MON-308 and MON-309 strains as well as that of the CD44 MON-37 clones ( Table S1 ) were compared to those of previously typed L . infantum and L . donovani strains of different zymodemes and different geographical origins ( Table 1 ) [16] , [17] , [18] , [21] . The population structure of the sample set was estimated using various methods , each one with different assumptions . Firstly , we employed the Bayesian genotype clustering program STRUCTURE v 2 . 3 . 1 [27] . When the K estimator derived from the second-order rate of change of the likelihood function with respect to K [28] was examined , a sharp signal was found at K = 2 ( Fig . 2A ) . This suggested that two major homogeneous gene pools shape the genetic structure of the analysed strains . At K = 2 , MON-1 strains form the first population and all non MON-1 strains , the previously characterized L . donovani strains and L . infantum non MON-1 as well as those from Turkey ( TR ) , were grouped in the second ancestral source population ( Fig . 2A ) . At K = 3 , the L . donovani population splits to form a distinct one ( Fig . 2A ) . Interestingly , at K = 4 the 6 strains from Çukurova region and the recently identified MON-37 isolates from Cyprus ( CY ) [12] , [13] , [16] group into a separate population , hereafter termed TR/CY non MON-1 . This population also included one of the MON-37 clones of the CD44 canine isolate ( CD44 cl . 1 ) . The EP59 strain could not be assigned to only one population but had shared membership to both the GR/TR MON-1 ( Greece/Turkey ) and TR/CY non MON-1 populations ( Fig . 2A ) . It is evident that in the previous estimation the uppermost hierarchical level of population structure was captured . However , the logarithm of probability of the data reached a plateau at K = 8 revealing the presence of additional subpopulations ( Fig . 2A ) . All populations were well defined and stable at K = 8 , and corresponded generally to the geographical origin of the strains . The eight main populations were: ( 1 ) IN1 ( India 1 ) , IN3 ( India 3 ) and LK ( Sri Lanka ) ; ( 2 ) SD/ET ( Sudan/Ethiopia ) ; ( 3 ) KE/IN2 ( Kenya/India 2 ) ; ( 4 ) L . infantum non MON-1 population 1; ( 5 ) L . infantum non MON-1 population 2; ( 6 ) TR/CY non MON-1 ( Turkey/Cyprus ) ; ( 7 ) SP/PT MON-1 ( Spain/Portugal ) ; and ( 8 ) GR/TR MON-1 ( Greece/Turkey ) . The main L . donovani complex genetic populations have been previously identified in various studies [16] , [17] , [18] , [21] and were also apparent in our analysis . Similarly , populations IN1 , LK , IN3 have been analysed extensively in other publications [16] , [21] and were also well-defined in our analysis when larger K values ( >10 ) were estimated ( not shown ) . The TR/CY non MON-1 population is maintained without change from K = 4 to K = 9 . However , at K = 10 it splits further to form the TR non MON-1 and the CY non MON-1 populations ( not shown ) . As suggested by Evanno et al . [28] , in order to find the hidden sub-group structure , the ancestral group subsets defined by the program should be reanalysed independently . As defined at K = 3 ( Fig . 2A ) , the analysis was repeated only for the ancestral non MON-1 population , which is comprised of L . infantum non MON-1 and TR/CY non MON-1 and excluded the well characterized L . donovani and L . infantum MON-1 populations . In the new analysis a strong ΔK peak was observed at K = 2 and a weaker one at K = 5 ( Fig . 2B ) . In addition , K = 5 was the most likely number of populations when the logarithm of probability of the data was plotted with respect to K . At K = 2 , L . infantum non MON-1 and TR/CY non MON-1 were the two observed populations while at K = 5 the TR/CY non MON-1 population splits into the TR non MON-1 and CY non MON-1 populations ( Fig . 2B ) . At this split the Cypriot CH32 strain ( Table 1 ) groups with the Turkish isolates and the CH34 strain shows a significant proportion of membership to the TR non MON-1 population ( Fig . 2B ) . The fact that similar results were observed when analysing all strains together and when focusing only on the non MON-1 strains corroborates this analysis to a great extent . Secondly , we applied the proportion of shared alleles ( Dps ) distance measure [30] to estimate population differentiation under the assumption of an Infinite Allele Model ( IAM ) mutation model . The matrix was calculated using the microsatellite profiles of all strains under study ( Table 1 ) , excluding the EP59 strain to avoid a false phylogeny due to its ‘hybrid’ profile . The generated midpoint rooted NJ tree is shown in Figure 3 . Four main clusters are observed corresponding to the 4 populations obtained by STRUCTURE at K = 4 . Furthermore , in accordance with STRUCTURE analysis ( Fig . 2B , K = 5 ) the TR/CY non MON-1 cluster is formed by two geographically defined subclusters , one containing all Turkish strains ( TR non MON-1 ) and the other containing the MON-37 Cypriot strains ( CY non MON-1 ) . Interestingly , the TR/CY non MON-1 clade is placed between a small group , which includes the strains MHOM/MT/85/BUCK , MHOM/IT/93/ISS800 , and the L . infantum non MON-1 clade . Confidence intervals were obtained by bootstrapping ( 1000 replicates ) . Although bootstrap support of >50% was observed between most subclusters , including that of the studied TR non MON-1 strains , it was weak for the basal nodes discriminating the major clusters . Factorial correspondence analysis ( FCA ) of the MLMT data corroborated our observations concerning the genetic relationships among the strains in our sample set and particularly the existence of a distinct population containing the strains from Turkey and Cyprus . The TR/CY strains were placed in an intermediate position between the MON-1 and the L . infantum non MON-1 populations ( Fig . 4 ) . Interestingly , the EP59 strain did not tightly group to the other TR non MON-1 strains but had a shifted position towards the MON-1 population , between the TR non MON-1 and the MON-1 populations . In addition , the BUCK strain ( Table 1 ) was placed very close to the TR non MON-1 strains ( Fig . 4 ) confirming previously described analyses . The expected heterozygosity ( He ) , as a measure of genetic diversity , was higher in the TR/CY non MON-1 strains ( 0 . 518 ) compared to MON-1 ( 0 . 276 for SP/PT , and 0 . 246 for GR/TR ) strains ( Table 2 ) . The TR/CY non MON-1 population's high genetic diversity was in fact comparable to that of the other L . infantum non MON-1 strain population ( 0 . 710 ) . This was also apparent from the greater genetic distances on the Neighbor-joining tree ( Fig . 3 ) . Either a single or two different alleles were observed for all strains in all the loci under study . More than two peaks , suggestive of aneuploidy or mixed heterozygous strains , did not occur . For all populations , the observed heterozygosity values were significantly lower than the expected ones and were higher in L . infantum non MON-1 ( 0 . 196 ) and TR/CY non MON-1 ( 0 . 180 ) strains compared to L . infantum MON-1 strains ( 0 . 031 for SP/PT and 0 . 019 for GR/TR ) ( Table 2 ) . The inbreeding coefficients of the TR/CY non MON-1 population was calculated at 0 . 662 , the lowest value amongst the 4 populations studied here ( FIS 0 . 890 for SP/PT MON-1 , 0 . 925 for GR/TR MON-1 and 0 . 732 for L . infantum non MON-1 ) . The highly positive FIS values can be due to the presence of different factors , such as population subdivision ( Wahlund effect ) or a high rate of gene conversion . The EP59 strain had an unusual high number of heterozygous loci with alleles characteristic for both the GR/TR MON-1 and TR/CY non MON-1 populations and was assigned to intermediate positions between these two populations ( Fig . 1 , 2 and 4 ) . Because this suggested that the EP59 strain might represent either mixed MON-1 and MON-308 strains or a MON-1/MON-308 hybrid we decided to clone EP59 . Four clones were obtained ( EP59 cl1 to EP59 cl4 ) and re-analysed by MLMT . Interestingly , all clones had identical MLMT profiles to the uncloned strain suggesting that it might be a hybrid ( Table S1 ) . We have also analysed one out of the three MON-37 CD44 clones , isolated in 2005 from a dog in Cyprus [12] . The three CD44 MON-37 clones ( CD44cl . 1-CD44cl . 3 ) gave the same size for the K26 amplicon ( 700 bp ) as the MON-37 strains whereas two other clones ( CD44cl . 4 and CD44cl . 5 ) gave an amplicon typical for MON-1 ( data not shown ) . Also , CD44 clones 1–3 presented identical microsatellite profiles and had allele sizes common to those of the MON-37 Cypriot population ( CY ) [12] , [16] . Interestingly these clones were heterozygous at 6 out of 14 loci studied ( Table S1 ) . This result confirmed preliminary MLMT data suggesting that the dog was co-infected with both MON-1 and MON-37 strains [12] .
In this study , we seek to further analyse strains of the L . donovani complex , isolated in Turkey from human CL foci in Çukurova ( CUK strains ) and from one human VL case in Kuşadasi ( strain EP59 ) , and to compare them with strains from Cyprus ( CH strains and the MON-37 CD44 clones ) . These Cypriot strains were typed previously as L . donovani MON-37 by MLEE but shown to be genetically distinct from the MON-37 strains of other regions [16] . In line with previous DNA-based data [6] , [19] , using the K26-PCR assay , the Turkish strains were inferred to the L . donovani complex . Interestingly , they did not present the K26 amplicon sizes corresponding to the L . infantum MON-1 zymodeme [25] and were thus characterized as non MON-1 . Subsequent MLEE analysis of four of the Turkish isolates revealed two novel L . donovani zymodemes , namely zymodemes MON-308 and MON-309 . The Turkish strains from Çukurova ( zymodeme MON-309 ) and the MON-37 strains from Cyprus differed in just two isoenzymes and hence MLEE suggested a close relationship between them . Notably , when the EP59 , CUK2 and CH35 strains ( a MON-37 strain from Cyprus ) were subjected to PCR-RFLP of the cpbEF gene [34] an L . donovani profile was revealed ( data not shown ) . Overall our data point out the need to validate PCR typing assays on a geographically representative panel of isolates for a given region and to account not only for intra-species variability but also for species identification . This applies especially to areas where different sympatric species are present , as in the EMR . Also we would like to emphasize that for regions with relatively limited data on species distribution of the so called L . donovani complex , the specificity of the PCR tests should be validated by including a sufficient number of reference strains that fall in the boundaries between L . donovani and L . infantum species . Otherwise DNA-based typing could be erroneous . The K26-PCR assay is a simple and quick method that could prove valuable for assessing simultaneously whether the causative agent of VL or CL belongs to the L . donovani complex and if it is a MON-1 zymodeme . The MLMT approach that is highly discriminatory and reproducible was chosen for accurately investigating the genetic diversity and population structure of the Turkish strains and for classifying them within the L . donovani complex . Considerable polymorphism , comparable to that observed within the L . infantum non MON-1 group was detected within the studied Turkish strains . This is demonstrated by higher He and Ho values ( Table 2 ) , long branches in the N-J tree ( Fig . 3 ) and a broad distribution of the strains in FCA ( Fig . 4 ) . The most striking outcome of the MLMT analysis was the identification of a ‘new’ main cluster within the L . donovani complex , proposing the existence of four main L . donovani complex populations . All non MON-1 Turkish isolates were grouped with the recently isolated MON-37 strains from Cyprus [12] , [13] in a single monophyletic group ( TR/CY non MON-1 ) of close relationship to the L . infantum MON-1 group ( Fig . 2–4 ) . The remaining three clusters , namely L . infantum MON-1 , L . infantum non MON-1 and L . donovani , are in agreement with previous publications [16] , [17] , [18] , [21] . The fact that the same main populations were identified when both model- and distance-based methods were applied for the analysis of the microsatellite data , confirms the validity of our analysis . The main subclusters in the N-J tree were also supported by bootstrap values greater than 50% . Basal notes of the major clusters had weak support but this is commonly observed because microsatellite data become less informative for distantly related taxa [30] . Furthermore , strains BUCK and ISS800 formed an additional small cluster at the base of MON-1 and TR/CY non MON-1 ( Fig . 3 ) . The strains also grouped in a subpopulation at K = 5 when analysed with STRUCTURE ( Fig . 2B ) and BUCK had a significant membership to the TR/CY non MON-1 population at K = 8 ( Fig . 2A ) . Grouping of ISS800 with CUK strains was also demonstrated by Svobodova et al . [6] . Several studies using different genetic markers have placed these two strains ( BUCK and ISS800 ) in ambiguous intermediate positions . Specifically , the strains were either assigned at the basis of the non MON-1 group right after the split from L . donovani [21] , [35] or close to the MON-1 group [18] , [36] , as observed in our analysis . The inclusion of the TR/CY non MON-1 strains seems to better clarify the ambiguous position of the two strains but still a larger set of strains should be analysed to reach a solid conclusion . Interestingly , in a previous work where the same strains from Cyprus were analyzed [16] it was shown that the MON-37 zymodeme is paraphyletic , encompassing strains with the same enzymatic profile but with different genotypes and consequently their origin still remained enigmatic . Here , we show that the Cypriot strains are very closely related to the Turkish strains . Therefore , the strains could have potentially been introduced into Cyprus from Turkey or vice versa . The geographic proximity of the countries and population movement between them further supports this scenario . Nevertheless , the possibility that these strains originate from neighbouring Middle East countries cannot be excluded . Parasites isolated from both human hosts ( EP59 , CUK1 ) and P . ( Larroussius ) tobbi sand fly vectors ( CUK2 , CUK3 , CUK4 , CUK7 and CUK10 ) clustered in the same MLMT groups , confirming that P . ( Larroussius ) tobbi is responsible for the transmission of these parasites in the Çukurova focus [6] . Apart from the proven vector P . ( Larroussius ) tobbi , other members of subgenus Larroussius could facilitate circulation of TR/CY non MON-1 parasites . Sand flies of this subgenus belong to permissive vectors [37] and are widely distributed throughout Middle East countries , including Cyprus , Greece and Turkey [6] , [13] , [38] , [39] . P . ( Larroussius ) perniciosus , a major L . infantum vector in the central and western Mediterranean , was recently shown to be highly susceptible for CUK strains and could readily transmit the CUK parasites by bite [40] . In Iran , six Leishmania strains isolated from P . ( Larroussius ) perfiliewi transcaucasicus were typed as L . donovani by cpb gene sequencing [11] . Notably , the strains from Çukurova were typed as L . donovani by MLEE , even though in our MLMT analysis they group between the L . infantum MON-1 and non MON-1 populations ( Fig . 2 and 3 ) . Discrepancies between MLEE typing and genotyping especially regarding the taxonomy of the “L . donovani” complex have already been described in a number of studies [2] , [36] , [41] , [42] . This is not so surprising since MLEE mainly distinguishes the species of the L . donovani complex on the basis of just the gene encoding glutamate-oxaloacetate transaminases ( GOT ) isoenzymes , EC 2 . 6 . 1 . 1 , which has been shown to have undergone convergent evolution [43] . Interestingly the putative hybrid EP59 strain was responsible for VL in a human patient from Western Turkey ( Kuşadasi ) . This intra-specific putative hybrid is most likely a product of genetic cross between strains of the MON-1 and the non MON-1 TR/CY populations . The occurrence of genetic exchange in the genus Leishmania had remained an elusive point until recently , when Akopyants et al . provided evidence that Leishmania promastigotes are capable of having a sexual cycle consistent with a meiotic process in the sand fly [44] . Also Sadlova et al . demonstrated experimentally hybridization of L . donovani in two vectorial species , P . perniciosus and Lutzomyia longipalpis [45] . Leishmania hybrids have already been identified between closely related species , such as L . braziliensis/L . panamensis [46] , [47] and L . major/L . arabica [48] , [49] . Intra-species Leishmania hybrids have also been observed in Algeria [50] , Tunisia [51] , Sudan [2] and Ethiopia [52] . Strains with mixed genotypes are also frequently observed [16] , [18] . Here , mixed genotypes were observed for strains CH32 and CH34 , which presented alleles of both TR non MON-1 and CY non MON-1 subpopulations ( Fig . 2B , K = 5 ) . Furthermore , the canine MON-37 clone from Cyprus ( CD44 cl . 1 ) had a shared membership to both the L . infantum non MON-1 and the CY non MON-1 populations ( Fig . 2B , K = 5 ) . Hybridization may allow adaptation to new ecological niches , vectors and hosts , including humans and domestic animals , and the efficient spread of new traits . Volf et al . [53] demonstrated that L . infantum/L . major hybrids could be transmitted efficiently by P . papatasi , a highly specific sand fly vector that has been shown to support only the development of L . major . Additionally , Nolder et al . [54] described the emergence and spread of L . braziliensis/L . peruviana hybrids with increased fitness in Peru . However , it is important to mention that the pathology of leishmaniasis varies and is determined not only by parasite genetics but also by other factors such as host genetics [55] . Parasites belonging to the TR/CY non MON-1 population have been isolated from two distant regions of Turkey , the south-eastern ( Çukurova ) and south-western ( Kuşadası ) parts , indicating that these parasites could be broadly distributed in Turkey . In line with this hypothesis , their presence in neighbouring countries also merits to be investigated . Parasites of the identified TR/CY population could be contributing to the generation of CL in this region . In this case , clinical management of these patients would require more accurate monitoring during drug treatment since these parasites could potentially cause VL , which can be lethal if left untreated . Furthermore , appropriate drug regimes should be administered considering different Leishmania species show significant variation in their sensitivity to drugs [56] . Distinguishing different L . donovani complex species/subspecies is therefore crucial for prognosis , drug regimes and proper disease control . In conclusion , our analysis indicates that the epidemiology of leishmaniasis in Turkey is more complicated than originally thought . Here we describe a new L . donovani sensu lato non MON-1 group of strains originating from Turkey and Cyprus , which can cause both CL and VL . The identified population is genetically different from all other L . donovani complex populations , including the L . infantum MON-1 population that is also present in these countries . Interestingly , a putative MON-1/TR/CY non MON-1 hybrid strain and some gene flow between the studied populations were also identified . The results described herein and those of previous studies showing that MLEE and DNA-based approaches for strain typing may lead to discrepant results , further support the call for a revision of Leishmania taxonomy [2] , [21] , [36] , [41] , [42] . This should be based on gene sequences , which are remarkably congruent and uncontroversial . Nevertheless , our findings are inconsistent with the current nomenclature of the L . donovani complex with just two species and argue in favour of recognizing a number of L . donovani subspecies . | In eastern Mediterranean , leishmaniasis represents a major public health problem with considerable impact on morbidity and potential to spread . Cutaneous leishmaniasis ( CL ) caused by L . major or L . tropica accounts for most cases in this region although visceral leishmaniasis ( VL ) caused by L . infantum is also common . New foci of human CL caused by L . donovani complex strains were recently described in Cyprus and Turkey . Herein we analyzed Turkish strains from human CL foci in Çukurova region ( north of Cyprus ) and a human VL case in Kuşadasi . These were compared to Cypriot strains that were previously typed by Multilocus Enzyme Electrophoresis ( MLEE ) as L . donovani MON-37 . Nevertheless , they were found genetically distinct from MON-37 strains of other regions and therefore their origin remained enigmatic . A population study was performed by Multilocus Microsatellite Typing ( MLMT ) and the profile of the Turkish strains was compared to previously analyzed L . donovani complex strains . Our results revealed close genetic relationship between Turkish and Cypriot strains , which form a genetically distinct L . infantum monophyletic group , suggesting that Cypriot strains may originate from Turkey . Our analysis indicates that the epidemiology of leishmaniasis in this region is more complicated than originally thought . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"biology"
] | 2012 | Multilocus Microsatellite Typing (MLMT) of Strains from Turkey and Cyprus Reveals a Novel Monophyletic L. donovani Sensu Lato Group |
Chagas disease is a tropical parasitic disease affecting about 10 million people , mostly in the Americas , and transmitted mainly by triatomine bugs . Insect vector control with indoor residual insecticides and the promotion of housing improvement is the main control intervention . The success of such interventions relies on their acceptance and appropriation by communities , which depends on their knowledge and perceptions of both the disease and the vector . In this study , we investigated school-aged children's knowledge and perception on triatomine vectors and Chagas disease to further understand how communities view this vector and the disease in Yucatan , Mexico . We performed an analysis of children's drawings on the theme of triatomines and their house in several rural villages , to explore in an open-ended manner their views , understanding and misconceptions . A total of 261 drawings were collected from children ages 6–12 from four villages . We found that children are very familiar with triatomine vectors , and know very well many aspects of their biology and ecology , and in particular their blood-feeding habits . On the other hand , their drawings suggest that the role of triatomines as vectors of a chronic and severe cardiac disease is less understood , and the main perceived health threat appears limited to the bite itself , as previously observed in adults . These results have important implications for the specific design of future education materials and campaigns , and for the promotion of the inclusion of children in raising Chagas disease awareness in these endemic communities .
Chagas disease is a parasitic disease representing a major public health problem in Latin America , with about 10 million infected people [1]–[3] . In Mexico , there are at least 1–2 million infected people , but the public health importance of the disease is still debated and Chagas disease remains a neglected disease [1] , [4] . The disease is caused by the protozoan parasite Trypanosoma cruzi , which is transmitted mainly by the feces of hematophagous bugs of the triatomine family , although secondary transmission mechanisms are increasingly contributing to the epidemiology of the disease [5] , and it is becoming more urbanized [6]–[9] . After a short acute phase , patients remain chronically infected for many years , in an asymptomatic stage , but 30–40% of them will progress to a chronic chagasic cardiomyopathy , and less frequently to the digestive form of the disease [10] . Control of the disease is mostly based on vector control with intra-domiciliary residual insecticides , sometimes associated with housing improvement , and treatment of infected patients with antiparasitic drugs benznidazole or nifurtimox [5] . A key aspect for the success of these interventions is their acceptance and compliance by the communities and patients , which is in turn affected by their knowledge and perceptions of the disease and the vector [11]–[15] . These socio-cultural aspects of Chagas disease have often been overlooked [16] . Therefore , a more integrative approach is needed as proposed in the ecosystem approach to health ( Ecohealth ) , which promotes the integration of ecological , biological and social aspects for a more effective and sustainable disease control within a context of social and economic development [15] , [17] , [18] . In a previous study in rural Mayan communities of Yucatan , Mexico , we found that communities had a good knowledge of triatomines and their habits , but most inhabitants had a limited understanding of the transmission mechanisms and clinical manifestations of Chagas disease [15] . Thus , they do not clearly perceive the disease ( and its vector ) as a serious health threat , in spite of a significant level of house infestation by triatomine vectors ( 15–54% of houses , [6] , [19]–[21] ) and a seroprevalence of T . cruzi infection of 1–5% in the population [22]–[24] . This lack of threat perception by the communities may negatively affect their participation in , and appropriation of disease control efforts . On the other hand , children are an understudied population , although they may provide a good representation of community knowledge and perceptions as well . Indeed , Piaget's initial view of children's health knowledge as limited has been challenged by other authors suggesting that they can develop complex mental representations to predict and understand important aspects of their environment [25] . They can also play a key role as health messenger , and thus are an excellent target for health education and the promotion of behavioral changes [26] . A number of different methods have been used to study children's knowledge and beliefs on a particular subject , and drawing analysis has been considered as a powerful tool to analyze children's imagery , as drawings are spontaneous and can reflect children's knowledge ( concepts and notions ) as well as popular culture and stereotypes [27] , [28] . The main advantages of using drawings is that children are used to express themselves graphically in schools , and drawing does not rely on verbal abilities , thus allowing greater freedom of expression by mitigating the limitations and biases which may be present in questionnaires and interviews [29] , [30] . Indeed , some concepts and representations can be difficult for a child to express clearly and explicitly in a verbal manner , but may be easily captured in a simple drawing . Thus , one of the main benefits of using drawings as a method to study children's knowledge is that it allows visualization of ideas and concepts in a non-constrained , non-pressured way [31] . Drawing analysis has thus been extensively used to assess children's knowledge and perceptions on a wide range of topics , including the environment , climate change , health education , and illness [30]–[34] . In this study , we investigated school-aged children's knowledge and perception on triatomine vectors and Chagas disease to further understand how communities view this vector and the disease in Yucatan , Mexico . We thus performed an analysis of children's' drawings on the theme of triatomine and their house in several rural villages , aiming to explore how children perceive triatomines , their feeding and resting habits , and their association with Chagas disease . We also evaluated how these views relate to those of adults , as well as to Chagas disease education materials and activities that have been used previously in the studied villages .
The study was carried out in the rural villages of Bokoba , Teya , Sudzal and Sanahcat , which are located about 15 to 20 km apart in the central part of the Yucatan state , Mexico . The four villages have very similar housing and socioeconomic conditions , as well as triatomine infestation levels [18] , and they have been part of a pilot project on integrated vector control for Chagas disease over the past few years [18] , [35] . As such , triatomine collection activities [18] , [36] , as well as serological surveys [24] , and some community education and awareness activities have been taking place in these villages , although the later only focused on adults [15] . Adults were found to be rather knowledgeable on triatomines , but did not associate them well with Chagas disease [15] . Children's drawings were collected through a drawing contest [37] , which was conducted in the four study villages between April and December of 2012 . Primary school students' aged 6–12 years old from each primary school ( six schools in total ) from all four villages were invited to participate in the contest . Variables collected included basic demographic data ( gender , age , village , name of school and grade ) . Formal instructions describing the contest and its requirements were presented in a pamphlet and a poster to the school officials and teachers . Students were asked to use a 50×65 cm poster size sheet of paper of any color for their drawings . Each student was given two weeks to create a drawing on the topic of “My house and triatomines” ( “Mi casa y el pic” , as triatomines are most commonly referred to as a “pic” in Mayan ) . Children were asked to produce the drawings in their own homes and bring them back to school . The following details were requested in the drawings: 1 ) To present representative aspects of their community , 2 ) To include triatomine bugs or “pic” ( local Mayan name for triatomines ) in their drawing , 3 ) To show where bugs hide inside/around the house , and 4 ) To show what/whom the bugs feed on ( see supplementary Text S1 ) . No further instructions were given , so that children could freely associate triatomines with any situation or characteristics , including or not Chagas disease . We conducted a content analysis of the drawings in order to examine the extent and context of children's knowledge about triatomine vectors and Chagas disease and to understand how previous vector control related activities in the villages have informed their knowledge and understanding . An award ceremony with exposition of the drawings was held in each village after completion of the study , to acknowledge the most informative and creative drawings . This event was used as a Chagas disease awareness event , which included local government officials , school teachers , local health providers and the research team . Drawings were used as an open-ended approach to explore children's own views and perspectives on triatomines and Chagas disease . Taking the social cognitive learning theory ( SCLT ) into consideration [38] , the aim of the analysis was to capture the diversity of the information communicated through drawings . This theory takes into account the relationship between individual factors , environmental influences , and observational learning and how they collectively influence knowledge and behavior [38] . It makes emphasis on social influence and on how external and internal social reinforcement influence learning and specific behaviors . Thus , children's ability to identify triatomines and depict their knowledge in a drawing format was considered heavily based on the social and environmental conditions that have shaped their perception and understanding . Because some drawings also included text messages or dialogues , both the image content of the drawings , as well as the text were analyzed . A total of 261 drawings were collected and given an ID number . The drawings were associated with basic demographic information of its author . Children's ages were divided into categories corresponding to 6–7 , 8–9 and 10–12 years old . To analyze the content of the drawings , a qualitative scoring tool was developed . General thematic categories were deductively defined based on established attributes of triatomines and Chagas disease of interest . These themes included triatomine description and activity , resting/hiding locations inside and outside the house , animal and human hosts , signs of disease/suffering and pathology descriptions . A number of inductively developed subthemes/codes emerged from a general review of the drawings . As the result , a total of 61 inductively and deductively defined codes were considered within the general themes . Each drawing was then scored by four independent scorers for the presence or absence of each of the 61 codes . The individual scoring results were combined to create consensus for each theme and subtheme . In case of major discrepancies among scorers , drawings were re-examined collectively , to reach a consensus score . Scores were then used to calculate the frequencies of the presence/absence of the different themes and subthemes in the drawings , and identify the most commonly represented features , as well as some unique features from some drawings . A Chi square statistical test was used to compare frequencies among age groups , gender , and villages using JMP 9 software . The text written on each drawing was transcribed into an excel sheet and then transferred to Nvivo 7 software for qualitative analysis to identify the most commonly used words as well as define the main themes . The messages were also analyzed to determine the potential source of information for the written text , as well as the type of messages written . The study was approved by both the World Health Organization and the Autonomous University of Yucatan institutional bioethics committees . Parents of the children provided written informed consent to participate in the study .
Each drawing had an average of 3 . 8 bugs drawn , and triatomine bugs could be clearly identified in 93% ( 243/261 ) of all drawings ( Figure 2 ) . In the large majority of drawings ( 71% , 186/261 ) , the adult stage of bugs was depicted , while in a minority of drawings nymphal stages were represented ( 7% , 19/261 ) . The majority of the triatomines in the drawings were scored as having medium detail ( 34% , 89/261 ) , but a large proportion of drawings also had triatomines presented with very precise details ( 22% , 58/261 ) , and 30% ( 78/261 ) were more difficult to identify ( Figure 2 ) . The 6–7 year old group had the largest percentage ( 40% ) of bugs drawn that could not be identified as triatomines ( Figure 2 ) . Nonetheless , many of the bugs that could not be readily identified as triatomines were depicted as blood feeding bugs . Thus , the overall quality of the triatomine drawings clearly indicated an excellent recognition and identification of the bugs by the large majority of children . A relatively small proportion ( 8% , 21/261 ) of the drawings depicted a night scene , corresponding to the nocturnal behavior of the triatomines , while the majority illustrated a scene in daylight ( 54% , 140/261 ) , and others were indistinguishable ( 38% , 99/261 ) . Analysis of the location of the bugs , both inside and outside the house , provided information on the perceived habitat and/or resting places of triatomines . The bugs were drawn outside of the house in 81% ( 212/261 ) of all drawings and inside in 32% ( 83/261 ) . The three most common areas outside the house where the bugs were located included trees ( 36% , 93/261 ) , in the grass ( 34% , 89/261 ) and on house walls ( 27% , 70/261 ) . Bugs were also shown on the fence walls of the yard ( made of piled rocks ) , firewood and/or dead trunks , as well as in chicken coops and animal corrals ( Figure 3 ) , which have all been described as potential refuges/habitats for bugs [39] . Inside the house , the most common area where the triatomine was drawn was in the bedroom and more precisely by the bed/hammock in 26% ( 68/261 ) of drawings , followed by the floor in 21% ( 54/261 ) of drawings . Triatomines were represented next to a light bulb in very few instances ( 1 . 5% , 4/261 ) ( Figure 3 ) . Most houses appeared to be made of concrete/cement ( 63% , 166/261 ) . Many drawings represented triatomines by a door or window of the house ( 29% , 76/261 ) , in agreement with the flying and invasive behavior of bugs in the region [15] , [18] , [40] . Some children associated triatomines with trash or junk piles in the yard , and others with water sources and puddles , suggesting some possible confusion with mosquitoes breeding sites [41] . In only 2% ( 6/261 ) of instances triatomines were portrayed flying , and most of the time appeared resting/hiding/feeding . In decreasing frequency , dogs , chicken , horses , pigs , cows and cats were depicted in the yards around the houses , with dogs and chicken also found inside the house in a few drawings ( 1% , 3/261 ) . Dogs ( 17% , 44/261 ) and chicken ( 13% , 33/261 ) were the most frequent domestic animals present . Triatomine bugs were clearly blood feeding on animal hosts in 15% ( 38/261 ) of the drawings , and portrayed next to or on the animal hosts in an additional 27% ( 72/261 ) of the drawings ( Figure 4 ) . This clearly illustrated the hematophagous behavior of triatomines . The most frequent animal feeding hosts for bugs were dogs ( 11/261 , 4% ) , followed by cows ( 3% , 9/261 ) and chickens ( 3% , 8/261 ) . Humans were present in 65% of drawings ( 170/261 ) , and mostly corresponded to children ( 32% , 84/261 ) . Importantly , triatomines were also described feeding on humans in many drawings ( 18% , 47/261 ) , and right next to humans in another 38% of drawings ( 98/261 ) , emphasizing an anthropophylic feeding habit of the bugs ( Figure 5 ) . Twenty six percent ( 68/261 ) of all drawings showed a human in bed or hammock and 15% ( 40/261 ) of all drawings depicted a sleeping person . A couple of drawings showed a triatomine feeding on flower nectar , and others associated them with water puddles . As an indication of a potential relationship between triatomines and a disease , facial and body expression of the humans were observed . Although in over half of the drawings the facial expression could not be determined , the humans were drawn with a positive , happy expression ( smiling , cheering ) in 29% of all drawings . In 13% of the drawings the human was visibly discontent and/or in pain and depicted frowning or crying ( Figure 5 and 6 ) . Animals were sometimes depicted as scared or in pain as well , when bugs were near them or feeding on them . The most frequent signs of disease depicted were bug bites in 17 drawings ( 7% ) , and rash in 10 drawings ( 4% ) ( Figure 6 ) . A sick person was present in 2 drawings . Cardiac disease was only rarely depicted ( Figure 6 ) . Some children also drew a health center or a doctor . Although no specific instructions were provided to the children regarding the inclusion of text in their drawings , over half of them included some textual message or dialogue . Theme analysis indicated that these texts referred to the location of triatomine bugs , their host feeding , general prevention and danger messages , and the association with ( Chagas ) disease . The most commonly occurring themes were those discussing the location of triatomines and their feeding on humans , and to a lesser extent on animals , in agreement with what was depicted in the drawings . Warnings and danger texts were also frequent , labeling triatomines as “Bad” , “Be careful” or “Dangerous” . Triatomine bites were often mentioned and in several instances associated with a severe or fatal disease , although the exact nature of the disease was rarely specified . In only a few cases , Chagas disease was mentioned , and in others , reference was made to heart disease . Some text messages were identical among several drawings , suggesting a collective activity . Other messages ( 8% , 22/261 drawings ) were directly copied from education materials distributed by our research group to inform adult participants of our vector control intervention on Chagas disease , and less frequently from materials provided by the Ministry of Health . These suggested some external inputs from a variety of sources in the drawings , either from parents at home or from the teachers in schools , rather than the children's own perceptions . Other routes of transmission were also mentioned a few times including transfusional and congenital . Surprisingly , oral transmission through the consumption of raw or insufficiently cooked meat from infected animals was mentioned in a few drawings . Because different Chagas disease-related activities had been occurring in each village , we also assessed potential differences in drawings among villages . Indeed , while there had been bug collections in all four villages , pilot vector control interventions had been only performed in Bokoba , Teya and Sudzal , and seroprevalence surveys of the inhabitants had been only performed in Teya and Sudzal . Overall , drawings from the village of Sanahcat had the highest proportion of drawings without triatomines clearly represented ( X2 = 19 . 3 , P = 0 . 022 ) . Bugs were also less frequently depicted inside the house ( X2 = 18 . 9 , P = 0 . 026 ) , as well as in a bedroom ( X2 = 16 . 3 , P = 0 . 012 ) and on/under a bed ( X2 = 8 . 7 , P = 0 . 033; Figure 7 ) . This suggested that in this village , bugs are more perceived as sylvatic/peridomestic , and not frequently entering houses . Also , drawings from Sanahcat presented significantly fewer representations of warning and danger images ( X2 = 9 . 1 , P = 0 . 027 ) , as well as fewer references to disease ( X2 = 8 . 4 , P = 0 . 038; Figure 7 ) . Associations of triatomines with water sources and puddles were only seen in drawings from the villages of Sanahcat and Sudzal . This suggested a lesser familiarity with triatomines of the children from Sanahcat . Nonetheless , in the villages of Sanahcat and Sudzal , bugs were more frequently depicted as blood feeding on humans compared to the other villages ( X2 = 35 . 5 , P<0 . 0001 ) . In Sanahcat , there were more associations of the bugs with bites ( X2 = 25 . 5 , P<0 . 0001 ) and rash ( X2 = 11 . 1 , P = 0 . 011 ) on human subjects , which were more rarely depicted in drawings from the other villages ( Figure 7 ) , suggesting a greater perception of triatomines as a health threat in Sanahcat . There were also some differences in the perceived habitats . Bugs were more frequently described in association with chicken coops and animal corrals in Sudzal ( X2 = 55 . 2 , P<0 . 0001 ) , although this may be a simple reflection of the overall higher abundance of domestic animals in the drawings from this village . In Sudzal and Teya , bugs were also more frequently located on yard fences compared to the other villages ( X2 = 21 . 6 , P<0 . 001 ) . Interestingly , in Teya bugs were much more frequently depicted near windows and doors ( X2 = 15 . 4 , P = 0 . 0015 and X2 = 9 . 5 , P = 0 . 023; Figure 7 ) , which seems clearly related to the pilot installation of window screens performed previously [35] . However , the other differences among drawings from the different villages could not be clearly associated with the nature of previous Chagas disease-related activities .
The knowledge and perceptions of communities related to insect vectors and the disease they transmit are crucial aspects which influence community acceptance of , and participation in , vector and disease control activities [15] , [17] . Indeed , communities need to appropriate the interventions for these to be self-sustained and effective . In that respect , children have been given little attention , even though they may provide a good representation of community knowledge and perceptions . They can also play the role of health messengers , and thus are an excellent target for health education and the promotion of behavior changes [26] , which is subsequently spread to other family members and friends . To our knowledge , our study is the first to focus on children's views of triatomine vectors and Chagas disease based on drawing analysis . A first striking observation was the large amount of information presented in the drawings and the attention to detail in many of them . This underlies the breadth and accuracy of children's understanding and knowledge of triatomines and their behavior , as well as indicates that children are very familiar with these bugs . Indeed , the unambiguous precision of many drawings clearly suggested that children have an intimate knowledge of the bugs based on their own observations and experiences . This is in agreement with previous observations of a rather elevated house infestation by triatomines in the region , and further highlights the risk of T . cruzi transmission to humans . Indeed , serological surveys have indicated a seroprevalence of 1–5% in the region [22] , [23] , and cases of children seropositive for T . cruzi have also been reported [24] . The greater familiarity with adult bugs compared to nymphal stages may also be a reflection of house infestation being mostly caused by invasive adult bugs , with limited colonization ( defined by the presence of nymphs ) , as reported previously [19] , [21] , [40] , [42] . Furthermore , several of the bug habitats and hiding places as well as blood feeding sources depicted by children , including firewood , rock piles and walls , artificial lights , chicken coops and dogs , have been identified as important determinants for house infestation by T . dimidiata in the region [6] , [18] , [43] . Thus , as observed in several other studies on children's drawings , analysis of their content is a powerful tool to explore children's views on a variety of issues including health [28] , [34] , and these studies consistently illustrate that children's knowledge and perceptions are much more extensive than we think or assume . For example , children have been found to be able to represent complex concepts related to AIDS [33] or genetics [25] , among other themes [27] , [29] , [31] , [44] , [45] . With respect to triatomines , their knowledge appears similar to , or possibly even more detailed than that we observed in adults from the same villages [15] . Nonetheless , a few incorrect depictions were also occasionally observed , such as triatomines breeding/hiding in water sources and puddles , as well as nectar feeding on flowers . These seem to have arisen from confusion with mosquitoes , and may be related to the extensive education and diffusion campaigns on mosquitoes as part of Dengue fever prevention activities . Indeed , current dengue control efforts are based in part on the identification and removal of mosquito breeding containers from backyards and homes , which is promoted by multimedia campaigns [41] . The idea of ( Chagas ) disease transmission by the consumption of raw/uncooked meat from infected animals is surprising as it seems to be rather rare . Indeed , while several Chagas disease outbreaks caused by the ingestion of foods contaminated by bug feces ( mostly fruit juice ) have been reported in South America in recent years [46] , there are only two suspected cases of infection by consumption of raw meat or blood from infected animals , suggesting that more studies are needed to establish the epidemiologic relevance of this mechanism of transmission in humans . The blood feeding of triatomines appears also well understood by the large majority of children , with several of the established feeding sources such as dogs and chickens [47]; Dumonteil et al . , unpublished data] well represented in their drawings . Nonetheless , as previously observed in adults [15] , there is a poor association of triatomines with a severe chronic disease , and with Chagas disease in particular . Most of their preoccupations are associated with the bug bite itself , which may cause pain or skin rash . This may be due to the difficulty in depicting a severe cardiac disease , even though the children appeared to have a high level of imagination and disposition to communicate , as evidenced by the danger images and the addition of creative text warnings in the drawings . Human expressions and emotions may also not be the best indicator of an association of bugs with disease , as the ability to depict and understand emotions is still undergoing development in school-aged children [44] , [45] , which may explain the large number of drawings with happy faces we observed . Nonetheless , the depiction of a severe disease , and of a cardiac pathology in particular , was rather uncommon . In comparison , in a study on genetics , children sometimes associated genetics with disease [25] , while in a drawing study on AIDS , the majority of children clearly depicted bedridden sick persons and death in their representations of disease long-term outcomes [33] . The lack of perception of triatomines as a vector of a severe cardiac disease may also reflect the invisible nature of the disease , of which they only have heard of within the social context of their communities ( i . e . observational learning as in SCLT [38] ) , compared to their more vivid experiences with triatomines . Further approaches should help clarify children's perceptions of Chagas disease . Taken together , these observations pinpoint to several specific aspects of Chagas disease that should be strengthened in future education and diffusion messages . First , it would be of key importance to take advantage of the excellent knowledge that children and adults have of triatomines to further build upon this knowledge in future materials . Second , the disease should be better explained , in terms of its symptoms , clinical manifestations and outcomes , to ensure that the severity and nature of Chagas disease is well understood by communities . Third , its relationship with triatomines , including the mechanisms of parasite transmission via triatomine feces upon blood feeding and the slow development of the disease , needs to be reinforced . A potential bias of our design was that children were given two weeks to complete their drawings as it was organized as a contest , while in most previous studies on children's drawings they were only given a much shorter time ( 30 min–2 h ) to realize their drawings on site . Their drawings may thus be seen as less spontaneous and more researched , but such contests have been used before with success to understand children's perceptions [37] . The fact that several children copied text messages from Chagas disease awareness materials clearly indicate that they did some research and/or had some potential input from parents or teachers . In any case , this should be seen positively as it demonstrates that these awareness materials are being rather extensively circulated , and more importantly socialized within the entire community , thereby effectively contributing to increased community knowledge and awareness on Chagas disease . However , not all the differences we observed among drawings from the different villages could be attributed to differences in the previous Chagas disease awareness and research-related activities that were performed in these villages , suggesting that there may also be intrinsic differences in communities' knowledge and perceptions of triatomines and Chagas disease . In conclusion , our exploration of children's knowledge and perceptions of triatomines and Chagas disease through drawings confirmed that this approach is a powerful tool to explore , in an open-ended manner , the views of children , their understanding and misconceptions of health-related topics . We found that children are very familiar with triatomine vectors , and know very well many aspects of their biology and ecology , and in particular their blood-feeding habits . On the other hand , their drawings suggest that the role of triatomines as vectors of a chronic and severe cardiac disease is less understood , and the main perceived health threat appears limited to the bite itself , as previously observed in adults . These results have important implications for the specific design of future education materials and campaigns , and the inclusion of children in promoting Chagas disease awareness in these endemic communities . | Chagas disease is a tropical parasitic disease affecting about 10 million people , mostly in the Americas , and transmitted mainly by triatomine bugs . The elimination of the insect vectors from houses is the main strategy for the prevention of the disease . The success of such interventions relies on their acceptance and appropriation by communities , which depend on their knowledge and perceptions of the disease and the vector . In this study , we performed the first analysis of school-aged children's views of triatomine vectors and Chagas disease in Yucatan , Mexico . We analyzed 261 drawings on the theme of “triatomines and my house” collected in four rural villages . Children appeared very familiar with triatomine vectors , but their role as vectors of a chronic and severe cardiac disease was less understood . These results can help design more specific education materials and campaigns , and promote the inclusion of children in raising Chagas disease awareness in these communities . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases",
"psychology",
"medicine",
"and",
"health",
"sciences",
"chagas",
"disease",
"communication",
"in",
"health",
"care",
"protozoan",
"infections",
"biology",
"and",
"life",
"sciences",
"vector-borne",
"diseases",
"collective",
"human",
"behavior",... | 2014 | Analysis of Children's Perception of Triatomine Vectors of Chagas Disease through Drawings: Opportunities for Targeted Health Education |
Numerous obesity loci have been identified using genome-wide association studies . A UK study indicated that physical activity may attenuate the cumulative effect of 12 of these loci , but replication studies are lacking . Therefore , we tested whether the aggregate effect of these loci is diminished in adults of European ancestry reporting high levels of physical activity . Twelve obesity-susceptibility loci were genotyped or imputed in 111 , 421 participants . A genetic risk score ( GRS ) was calculated by summing the BMI-associated alleles of each genetic variant . Physical activity was assessed using self-administered questionnaires . Multiplicative interactions between the GRS and physical activity on BMI were tested in linear and logistic regression models in each cohort , with adjustment for age , age2 , sex , study center ( for multicenter studies ) , and the marginal terms for physical activity and the GRS . These results were combined using meta-analysis weighted by cohort sample size . The meta-analysis yielded a statistically significant GRS × physical activity interaction effect estimate ( Pinteraction = 0 . 015 ) . However , a statistically significant interaction effect was only apparent in North American cohorts ( n = 39 , 810 , Pinteraction = 0 . 014 vs . n = 71 , 611 , Pinteraction = 0 . 275 for Europeans ) . In secondary analyses , both the FTO rs1121980 ( Pinteraction = 0 . 003 ) and the SEC16B rs10913469 ( Pinteraction = 0 . 025 ) variants showed evidence of SNP × physical activity interactions . This meta-analysis of 111 , 421 individuals provides further support for an interaction between physical activity and a GRS in obesity disposition , although these findings hinge on the inclusion of cohorts from North America , indicating that these results are either population-specific or non-causal .
Obesity is a major risk factor for many non-communicable diseases including type 2 diabetes , cardiovascular disease , and certain cancers [1] . Genetic predisposition and lifestyle factors are known to increase obesity susceptibility , and the technological breakthroughs that came with genome-wide association studies ( GWAS ) have led to the successful identification of a large number of obesogenic loci [2]–[6] . Recent studies suggest that physical activity may modify genetic susceptibility to obesity , with the genetic burden being higher in physically inactive compared with active persons [7]–[9] . The most extensively studied example of a gene × physical activity interaction in obesity is for the FTO locus [7] , [10] , which was recently replicated in a meta-analysis comprising 240 , 000 persons [11] . Elsewhere , Li et al reported that physical activity offsets the aggregated genetic risk of 12 obesogenic loci [12] . In the current study , we aimed to replicate the findings of Li et al [12] in a sample collection of 111 , 421 individuals of European ancestry . We also undertook detailed analyses focused on the role of within- and between-study factors to establish how the design of gene × environment interaction meta-analyses impacts the power to detect interactions .
The forest plot in Figure 1 shows the interaction coefficients across the 11 cohorts included in the meta-analysis , along with the overall interaction effect estimate ( Pinteraction = 0 . 015 ) . Table 1 summarizes the adjusted main effects of the GRS on BMI and obesity in the combined data from all cohorts and by strata of physical activity . Each unit increase in the GRS , equivalent to one BMI-raising allele , was associated with a mean 0 . 161 ( SE = 0 . 006 ) kg/m2 higher BMI ( P = 2 . 1×10−176 ) , which corresponds to 465 g heavier weight for a person 1 . 70 m tall . Overall , among physically inactive individuals ( with a Cambridge Physical Activity Index [CPAI] of 1 ) , each additional BMI-raising allele was associated with 0 . 186 ( SE = 0 . 006 ) kg/m2 higher BMI , equivalent to 538 g in weight for a person 1 . 70 m tall ( P = 4 . 8×10−47 ) , whereas the effect in the most physically active group ( CPAI of 4 ) was 0 . 143 kg/m2 per GRS allele ( SE = 0 . 011 , P = 5 . 6×10−40 ) , or 413 g in weight for a person 1 . 70 m tall . In the ‘combined active’ group ( individuals with a CPAI of 2–4 ) , each additional risk allele was associated with 0 . 150 kg/m2 ( SE = 0 . 007 , P = 3 . 3×10−107 ) higher BMI , or 434 g in weight for a person 1 . 70 m tall ( Figure 2 ) . As illustrated in Figure 3 , in the inactive group ( CPAI of 1 ) , the difference in BMI between persons with a low ( ≤11 alleles ) and high ( >11 alleles ) GRS was 0 . 647 kg/m2 ( SE = 0 . 06; P = 1 . 9×10−25 ) , while the difference in the combined active group was 0 . 532 kg/m2 ( SE = 0 . 03; P = 6 . 6×10−67 ) . The CPAI characterizes total physical activity levels by considering both occupational and leisure time physical activity [13] . Sensitivity analyses were performed in the GLACIER and MDC cohorts ( n = 39 , 000 ) where interaction terms ( gene × physical activity ) were modeled separately for occupational and leisure time physical activity , but these results were not materially different from the main analyses ( data not shown ) . Within these two cohorts , we additionally adjusted the models for putative confounding by smoking and education , but the results were essentially the same irrespective of whether these additional covariates were or were not included; hence , for the sake of comparability , we focus on the results with the regression models adjusted as reported by Li et al [12] . We also undertook sensitivity analyses in European and North American cohorts separately ( Supplementary Figures S1a and S1b ) , which revealed a statistically significant GRS × physical activity interaction effect in the latter ( n = 39 , 810 , Pinteraction = 0 . 014 ) , but not the former ( n = 71 , 611 , Pinteraction = 0 . 275 ) . In analyses modeling the interaction of each of the 12 individual SNPs and physical activity , two tests of interaction were nominally statistically significant: the FTO rs1121980 variant , which concurs with previous reports of interaction at this locus [11] , and the SEC16B rs10913469 locus , which has not previously been reported ( Table 2 ) . It should be noted that several of the cohorts used here are included in Kilpeläinen et al . [11] , and so this is not entirely independent confirmation of these findings . The magnitude of the interaction effects ( βGE ) for FTO rs1121980 and SEC16B rs10913469 variants was −0 . 052 and −0 . 049 kg/m2 per risk allele respectively , which compares with βGE of −0 . 108 kg/m2 per 8 . 33 alleles for the GRS ( equivalent to 1 allele on the bi-allelic scale ) . For FTO , the interaction effect was almost 10-fold larger in North American than in European cohorts , whereas for the SEC16B locus , the interaction effect was approximately twice the magnitude in North American vs . European cohorts . Supplementary Table S2 shows individual SNP interaction results across each of the 11 cohorts . In models excluding the FTO and SEC16B variants from the GRS , the interaction test was no longer statistically significant ( in the entire cohort [Pinteraction = 0 . 25] or separately within the cohorts from North American [Pinteraction = 0 . 39] and Europe [Pinteraction = 0 . 44] ) , strongly suggesting that the GRS × physical activity interaction result is driven by the inclusion of one or both of these variants .
Here we sought to replicate a widely cited study in which an interaction on BMI was reported between physical activity and a GRS comprised of 12 obesity-predisposing gene variants [12] . The original study is one of the largest and most well conducted single-cohort interaction studies published to date , yet to our knowledge no evidence has been published to show that these findings are replicable . Our study included a collection of cohorts whose sample totaled almost six times the size of the study reported by Li et al [12]; the meta-analyzed interaction coefficient is directionally consistent with the original report [12] and statistically significant in the current analysis ( Pinteraction = 0 . 015 ) . In secondary analyses , we explored whether any of the individual SNP × physical activity interaction tests were statistically significant; of these , the FTO locus ( rs1121980 ) ( Pinteraction = 0 . 003 ) , consistent with previous findings [11] , and the SEC16B rs10913469 variant yielded statistically significant interaction effects ( Pinteraction = 0 . 025 ) . The latter finding was not statistically significant after correction for multiple testing , there is no published literature suggesting that this locus is exercise-responsive , and a recent analysis in a randomized clinical trial of lifestyle intervention did not yield evidence of SNP × treatment interactions at the SEC16B rs10913469 locus on weight change phenotypes [17] , although that analyses was likely underpowered and may be false negative . Thus , validation of the interaction effect observed here for SEC16B rs10913469 is necessary to confirm or refute its effect-modifying role for physical activity and obesity . It is widely acknowledged that initial reports of genetic association signals are often of considerably greater effect magnitude than yielded by subsequent replication attempts; this phenomenon is termed the Winner's curse [18] . The large Winner's curse differential ( ΔβGE = 0 . 057 kg/m2 per GRS allele for the comparison of βGE reported by Li et al [12] and observed in the current study ) has a dramatic effect on the sample size required for replication , with around 530 , 000 individuals ( >25 times the size of the original study ) being required to yield power of 80% to detect the interaction effect reported in this study ( βGE = −0 . 013 kg/m2 per GRS allele ) . We also conducted a range of simulation analyses to determine how within and between study factors impact power to detect interactions in meta-analyses . We show that the optimal setting is one where i ) for a given interaction effect size ( βGE ) , the independent variables are expressed on a continuous scale ( and if physical activity is dichotomized and the interaction effect is approximately larger the categories should be equally prevalent ( i . e . , 50%/50% ) ) , ii ) the variance in the GRS is large , iii ) the GRS and environmental exposure are correlated , and iv ) the population variance in the outcome is small , which in part relates to whether exposure and outcome measurements are standardized across studies and measured with reasonable precision ( the latter of which is discussed at length elsewhere [16] ) . One of the principal arguments for conducting and reporting studies on gene × lifestyle interactions is that they may help identify persons within target populations who are likely to respond well or poorly to specific lifestyle interventions , thus optimizing the delivery and success of the interventions; the same principle may apply to other medical therapies such as drug treatment and surgery . The targeting of lifestyle interventions using genetic information is appealing as it may improve cost-efficiency , reduce harmful side effects , and increase the health-promoting effects of diet and lifestyle factors [19] . However , very few reported gene × lifestyle interactions have been replicated , which may be because many of the original findings were false positive , the reported interaction effects were cohort-specific , or because subsequent studies were underpowered and yielded false negative results [20] . The study by Li et al [12] appears well conducted and was performed in a relatively large cohort . The paper was also published in a high impact general medical journal , which implies that the authors' findings are clinically relevant , yet , like most studies of gene × environment interaction , they lacked replication . Importantly , the clinical translation of findings on gene × lifestyle interactions requires that the interaction effect sizes are of a sufficient magnitude to ensure that stratified therapeutic interventions will yield meaningfully different results across genotype groups . The interaction effect size reported in this study is probably too small to be of any clinical value; it is worth noting , though , that in observational studies , where the precision and accuracy with which exposures and outcomes are measured is often low , and where synthetic genetic associations exist ( i . e . , the observed locus is merely a tag for the latent functional locus ) , the underlying interaction effect sizes are likely to be underestimated . A second incentive for conducting studies on gene × lifestyle interactions is that doing so may elucidate biological pathways that lead to the targeting of therapeutic interventions . Most or all of the SNPs studied here probably tag functional variants , with no specific functional role of their own . The functional relevance of the genes most proximal to these SNPs is discussed in detail elsewhere [2]–[6] . The majority of these genes regulate CNS-mediated body weight regulation , energy balance , taste , and satiation [21]; although not clearly established , these genes might also regulate reciprocal behaviors; for example , variants in MC4R [22]–[24] and FTO [25] , [26] are reportedly associated with physical activity . Although we found statistical evidence of an interaction between physical activity and the GRS in the meta-analysis , it is unlikely that all of the gene variants that comprise the GRS contribute to this interaction effect . For example , the FTO variant included in the GRS has been shown previously to interact with physical activity on obesity [11] , a finding that was confirmed here , and the SEC16B variant also yielded a nominally significant interaction effect in this study . In combination , the two variants yielded an interaction effect size comparable to that seen here for the GRS × physical activity interaction , and the GRS × physical activity interaction test was not statistically significant when the FTO and SEC16B variants were excluded from the GRS , suggesting that these two loci underlie the aggregate genetic effect of all 12 SNPs combined . It is difficult to accurately speculate on whether the GRS × physical activity interaction reported by Li et al [12] is also driven by the FTO and SEC16B interaction effects , as formal comparisons of this nature were not reported in their paper . Refitting the alleles that comprise a GRS to maximally exploit this information in a regression model ( i . e . , by weighting the alleles by their interaction effect estimates obtained from SNP × physical activity interaction analyses ) would likely increase the magnitude of the observed interaction effect for the GRS; however , to achieve this with minimal bias would require further sample collections to validate these new genetic models , which goes beyond the scope of the present study . Nonetheless , we include the relevant information in Table 2 , so that other investigators can construct such weighted models . It is also important to highlight that the interaction results reported by Li et al [12] were not statistically significant once persons with prevalent CVD and cancer were excluded; the inclusion of these individuals may have confounded the interaction effect owing to reporting biases attributable to disease labeling or changes in weight and behavior attributable to the disease processes , although the fact that we have replicated their findings in cohorts that were largely free of these diseases suggests this is not the case . It is also possible that the inclusion of diseased individuals in Li et al's study [12] augmented the interaction effect through hitherto unknown causal mechanisms . As a general point , it is important to bear in mind that in observational studies , such as those reported here , marginal and interaction effect estimates may not reflect causal processes . This is because physical activity and obesity correlate with other lifestyle , sociodemographic , and metabolic factors , and the gene variants included in the GRS are unlikely to be functional . Thus , even replicated examples of gene × lifestyle interactions may be confounded by latent variables . Reverse causality is a further concern , particularly with cross-sectional data ( for example , it is possible that there is a relationship between the GRS and physical activity that is dependent on BMI level ) . In summary , our meta-analysis of 111 , 421 samples from 11 cohorts of European ancestry yielded results that support those of Li et al [12] . However , these effects appear evident only when the cohorts from North America ( n = 39 , 810 ) are included in this meta-analyses . We also demonstrate using simulated data that combining many small cohorts that vary in their classification of physical activity and other factors is a relatively inefficient approach to studying interactions; hence , future studies of gene × lifestyle interactions might prove most effective if focused on a small collection of large cohorts within which standardized and valid lifestyle assessment methods are available .
A total of 111 , 421 participants from the 11 participating cohorts had genotype and phenotype data necessary for the current analyses . Descriptions of the cohorts included in the current analyses are shown in supplementary Table S6 . All participants provided written informed consent and the studies were approved by the relevant institutional review boards and conducted according to the Declaration of Helsinki . In most studies , height and weight were measured using wall-mounted stadiometers and calibrated balance-beam scales , respectively ( See Supplementary Table S7 ) . By exception , weight for the NHS , HPFS [27] , and WGHS [28] were self-reported . BMI was calculated as weight in kilograms ( kg ) divided by height in meters squared ( m2 ) . Obesity was defined according to WHO criteria [29] . Information on physical activity was obtained from self-administered questionnaires , which in most instances were validated . Occupational physical activity in most studies was categorized as i ) sedentary or standing; ii ) light but partly physically active; iii ) light and physically active; and iv ) sometimes or often physically straining . Leisure time physical activity during the past three months was categorized as exercising: i ) occasionally; ii ) 1–2 times/week; iii ) 2–3 times/week; or iv ) >3 times/week . Among leisure-time physical activity ( four categories ) , participants with missing information were given the lowest intensity score , i . e . classified as being ‘occasionally active’ . The CPAI was computed by cross-tabulation of occupational and leisure time physical activity , classifying an individual's total physical activity level according to a four-level scale ( inactive , moderately inactive , moderately active and active ) , as previously described [13] . Because some cohorts could not compute the CPAI owing to a lack of specific physical activity data , a binary variable was computed in all cohorts , which classified participants into active ( top 80% of the physical activity frequency distribution ) and inactive ( bottom 20% of the physical activity distribution ) . This classification most closely matches the frequency distribution obtained when dichotomizing the CPAI variable by combining moderately inactive , moderately active and active individuals ( see Supplementary Table S7 for further details ) , but , as noted in the Results , may not be the most statistically powerful classification . DNA was extracted from peripheral blood cells and diluted using standard approaches ( see Supplementary Table S8 for further details ) . Twelve established obesity susceptibility loci [2]–[6] ( or their proxies with an r2>0 . 8 ) were genotyped in the 11 cohorts ( Supplementary Table S8 ) . In all cohorts , the genotyping success rates for all 12 variants exceeded 95% and most genotypes were in Hardy-Weinberg equilibrium ( P>0 . 001 ) . The exception to this was for the SH2B1 rs7498665 SNP in the METSIM and HEALTH2006 cohorts , which did not conform to Hardy Weinberg expectations; sensitivity analyses indicated that removing this SNP from the GRS for the METSIM cohort made no material difference to the overall results ( data not shown ) , and so the results shown here are for the full GRS . At each SNP locus , genotypes were coded as 0 , 1 and 2 indicating the number of risk alleles ( those associated with higher BMI in previous meta-analyses [2]–[6] ) and the overall genetic burden for each participant was determined by summing the total number of risk alleles into a GRS , using methods previously described [30] . In cohorts where genotypes were directly assessed ( i . e . , not imputed from GWAS data ) , missing genotypes were imputed in participants with four or fewer missing values using previously described methods [31] . Sensitivity analyses performed in the GLACIER and MDC cohorts ( n = 39 , 000 ) showed that there was no material difference in the effect estimates when analyses were performed with or without imputed genotypes ( data not shown ) , so here only results for the GRS using imputed values are presented . The GRS was normally distributed in all cohorts . Statistical analyses were performed using the SAS software ( SAS Institute , Cary , NC ) , R software ( http://www . r-project . org/ ) and STATA ( version 12 , StataCorp , College Station , TX , USA ) . General linear models ( GLM ) were used to test the association of the GRS with BMI . Logistic regression was used to test genetic associations with obesity . All analyses were adjusted for age , age2 , sex , study center ( for multi-center studies ) , and physical activity ( where appropriate ) , and we assumed additive effects of the alleles . Interaction tests for individual SNPs and the GRS with physical activity ( for outcomes BMI or obesity ) were performed by including a SNP ( or GRS ) × physical activity interaction term in the model , with the marginal effect terms also included . The genetic effect estimates for BMI were also calculated by strata of physical activity ( i . e . inactive vs . combined active ) , as described above . Meta-analyses were undertaken using the metan command in STATA ( version 12 , StataCorp , College Station , TX , USA ) . A summary interaction effect estimate was calculated for all 11 cohorts combined using meta-analysis weighted by cohort sample size to summarize the pairwise ( SNP/GRS × physical activity ) interaction coefficients and SE derived from each cohort . Meta-analyses were repeated using random and fixed effects models , but between-study heterogeneity was low ( χ2 = 15 . 51 , I2 = 3 . 3% and P-val = 0 . 415 ) ; thus , the results were not materially different to the weighted approach ( data not shown ) , leading us to present only the weighted results here . Analysis of data from the InterAct Study , which includes multiple sub-cohorts , was conducted as described elsewhere [32] . The full InterAct Study includes two Swedish study centers in Malmö and Umeå , which overlap extensively with the GLACIER and MDC cohorts . Thus , these Swedish InterAct cohort samples were not included in the main analyses . The code-generating program mlPowSim [33] was used to generate R code for simulations and power estimation with 1 , 000 iterations for each sample size simulation . In order to estimate power for different samples sizes , we simulated a 12 SNP GRS using a random normal distribution with mean ( s . d . ) 11 . 2 ( 2 . 2 ) ; physical activity was simulated using a binomial distribution assuming the population prevalence of physical inactivity was 30% , as estimated by Li et al . The approach ( described in detail in the Supplementary Material S1 ) was used to simulate different scenarios for the predictor variables: i ) with the GRS expressed as a continuous or dichotomized variable ( Supplementary Figures S2a and S2b ) , ii ) a range of frequencies for the binary physical activity variable and variances ( σ2 ) ( Figure S3 , iii ) a range of effect sizes for βGE ( Supplementary Figures S2a and S2b ) , iv ) a range of covariances between the two predictor variables ( Figure S3 ) , and v ) a range of variances ( σ2 ) for the population ( Supplementary Table S5 ) . The main power calculations were performed using estimates obtained from Li et al [12]: a GRS marginal effect ( βG ) of 0 . 154 kg/m2 per GRS risk allele and a physical activity marginal effect ( βE ) of −0 . 313 kg/m2 ( active vs . inactive ) , physical inactivity prevalence of 30% , and s . d of ±3 . 5 . We assumed that the GRS and physical activity are not correlated and a two-sided critical alpha of 0 . 05 was used in the calculations . Although the interaction effect estimate ( βGE ) is not explicitly reported in Li et al's paper , we were able to estimate this from the GRS effect estimates reported in Table 2 of their paper ( βGE∼−0 . 07 ) by approximating the difference of βG between the two combined activity categories ( active vs . inactive ) . To accommodate imprecision in the estimation of βGE and the possibility that Li et al's study [12] was affected by the ‘winner's curse’ [18] and thus over-estimated the interaction effect size one could hope to observe in other cohorts , we show statistical power estimations for interaction effects ranging from −0 . 05 to −0 . 10 ( Supplementary Figure S2a ) . We also simulated the GRS as a binary variable and compared power using this approach with one where the GRS is expressed on a continuum ( Supplementary Figure S2b ) , as GRSs are often reported on the binary scale in genetic association studies . | We undertook analyses in 111 , 421 adults of European descent to examine whether physical activity diminishes the genetic risk of obesity predisposed by 12 single nucleotide polymorphisms , as previously reported in a study of 20 , 000 UK adults ( Li et al , PLoS Med . 2010 ) . Although the study by Li et al is widely cited , the original report has not been replicated to our knowledge . Therefore , we sought to confirm or refute the original study's findings in a combined analysis of 111 , 421 adults . Our analyses yielded a statistically significant interaction effect ( Pinteraction = 0 . 015 ) , confirming the original study's results; we also identified an interaction between the FTO locus and physical activity ( Pinteraction = 0 . 003 ) , verifying previous analyses ( Kilpelainen et al , PLoS Med . , 2010 ) , and we detected a novel interaction between the SEC16B locus and physical activity ( Pinteraction = 0 . 025 ) . We also examined the power constraints of interaction analyses , thereby demonstrating that sources of within- and between-study heterogeneity and the manner in which data are treated can inhibit the detection of interaction effects in meta-analyses that combine many cohorts with varying characteristics . This suggests that combining many small studies that have measured environmental exposures differently may be relatively inefficient for the detection of gene × environment interactions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"systems",
"biology",
"medicine",
"nutrition",
"clinical",
"epidemiology",
"clinical",
"research",
"design",
"genetic",
"association",
"studies",
"epidemiology",
"obesity",
"cohort",
"studies",
"genetics",
"population",
"genetics",
... | 2013 | Gene × Physical Activity Interactions in Obesity: Combined Analysis of 111,421 Individuals of European Ancestry |
Odor-guided behaviors , including homing , predator avoidance , or food and mate searching , are ubiquitous in animals . It is only recently that the neural substrate underlying olfactomotor behaviors in vertebrates was uncovered in lampreys . It consists of a neural pathway extending from the medial part of the olfactory bulb ( medOB ) to locomotor control centers in the brainstem via a single relay in the caudal diencephalon . This hardwired olfactomotor pathway is present throughout life and may be responsible for the olfactory-induced motor behaviors seen at all life stages . We investigated modulatory mechanisms acting on this pathway by conducting anatomical ( tract tracing and immunohistochemistry ) and physiological ( intracellular recordings and calcium imaging ) experiments on lamprey brain preparations . We show that the GABAergic circuitry of the olfactory bulb ( OB ) acts as a gatekeeper of this hardwired sensorimotor pathway . We also demonstrate the presence of a novel olfactomotor pathway that originates in the non-medOB and consists of a projection to the lateral pallium ( LPal ) that , in turn , projects to the caudal diencephalon and to the mesencephalic locomotor region ( MLR ) . Our results indicate that olfactory inputs can induce behavioral responses by activating brain locomotor centers via two distinct pathways that are strongly modulated by GABA in the OB . The existence of segregated olfactory subsystems in lampreys suggests that the organization of the olfactory system in functional clusters may be a common ancestral trait of vertebrates .
Olfactory cues can trigger goal-directed locomotor behaviors , such as homing , predator avoidance , or food and mate searching [1–11] . It is only recently that the neural pathways and mechanisms involved in transforming olfactory inputs into locomotor behavior were characterized for the first time in a vertebrate species , the lamprey [12 , 13] . It consists of a specific neural pathway extending from a single glomerulus located in the medial part of the olfactory bulb ( medOB ) to the mesencephalic locomotor region ( MLR ) , with a relay in the posterior tuberculum ( PT ) [12] . In all vertebrates , the MLR acts as a motor command center that controls locomotion via descending projections to brainstem reticulospinal ( RS ) neurons [14–22] . This olfactomotor pathway is present throughout the life cycle of lampreys , whether in larvae , newly transformed , parasitic , or spawning animals [12] . Yet , olfactory-induced motor behaviors can be life stage specific in lampreys . For instance , at the parasitic stage , lampreys feed on fish that they detect using olfactory cues [23] . Then , when sexually mature , the adults are attracted upstream by migratory pheromones released by larvae [24–26] . Once upstream , the females are attracted to males by sex pheromones [27 , 28] . The general organization of the lamprey olfactory system , from the periphery to the central nervous system ( CNS ) , is very similar to that of other vertebrates . The peripheral olfactory organ is composed of a main olfactory epithelium and an accessory olfactory organ [29–31] . Axons from olfactory sensory neurons ( OSNs ) of the olfactory epithelium terminate in the olfactory bulb ( OB ) . As in other vertebrates , the OB can be divided in two subregions , based on their inputs . The main olfactory bulb ( MOB ) , which occupies the whole OB except its medial part ( i . e . , the medOB ) , receives inputs from the main olfactory epithelium . The medOB , on the other hand , receives inputs from OSNs located in the accessory olfactory organ [32–34] . The OB of vertebrates constitutes the primary olfactory center of the CNS and , as such , filters and actively shapes sensory inputs to secondary olfactory structures [35 , 36] . This processing of sensory inputs in the OB is driven by modulatory inputs coming from the numerous neurotransmitter systems present in the OB of vertebrates [37 , 38] . GABA is the main inhibitory neurotransmitter in the CNS , and numerous GABAergic processes are present in the OB of several vertebrate species [39–43] . GABAergic neurons of the OB are believed to play a critical role in olfactory processing by providing inhibition to the bulbar microcircuitry [44] . However , their effect on the outputs of the OB and ultimately on behavior is far less understood . Here , we hypothesized that GABAergic neurons of the OB could play a significant role in modulating transmission in the olfactomotor pathway of lampreys . To address this , we used anatomical ( tract tracing and immunohistochemistry ) and physiological ( intracellular recordings ) techniques .
The present study showed abundant GABAergic cell bodies and processes in the OB ( n = 10 adult animals , Fig 1 ) , thus confirming the findings of Meléndez-Ferro and colleagues [43] . GABAergic neurons were mainly observed in the central region of the OB ( internal cell layer [ICL] , Fig 1A and 1D and S1 Fig ) , where the most common OB interneuron type , the granule cell , was described [45] . GABAergic processes were found all over the OB , including in and around the glomeruli of both the MOB ( Fig 1A and 1C ) and the medOB ( Fig 1A and 1B and S2 Fig ) . To investigate the physiological role of the GABAergic circuitry in the OB , local microinjections of the GABAA receptor antagonist , gabazine , were made into restricted areas of the OB , while stimulating the olfactory nerve ( ON ) and intracellularly recording from RS neurons on the same side of the brain . Gabazine injections ( 0 . 1 mM , 1 . 4 ± 1 . 6 nL ) in the medOB ( n = 60 synaptic responses; n = 6 neurons; n = 6 larval animals; Fig 2 ) were found to amplify synaptic responses of RS neurons to electrical stimulation of the ON ( amplitude increase of 372 . 2 ± 277 . 5%; p < 0 . 05; no statistical differences between control and washout; Fig 2B and 2C ) . RS neurons from all four reticular nuclei ( mesencephalic reticular nucleus and anterior , middle , and posterior rhombencephalic reticular nuclei ) responded similarly as shown by calcium imaging experiments ( n = 362 neurons; n = 6 adult animals , S3 Fig ) . Extracellular recordings of the OB further showed that responses of OB neurons to ON stimulation were greatly increased under gabazine ( n = 60 responses; n = 6 animals , S4 Fig ) , thus corroborating our previous findings . In addition to increasing the responses of RS cells , stimulation of the ON after gabazine injection in the medOB even induced motor discharges in the ventral roots . The neural activity consisted of rhythmic discharges alternating on both sides , a hallmark of fictive swimming ( in 58 . 1% of trials; n = 36 locomotor bouts out of 62 trials for gabazine versus 0 out of 78 for control; n = 9: three adult animals and six larval animals , Fig 3 and S1 Data ) . Because the density of GABAergic processes seemed relatively similar in the medOB and the MOB , we hypothesized that the neural activity in the MOB could be modulated by GABA , as observed for the medOB . To test this hypothesis , the effect of gabazine injections in the MOB on RS cell responses was examined . As shown for the medOB , gabazine injections into the MOB ( 0 . 1 mM , 1 . 8 ± 2 . 0 nL ) enhanced the RS neuron responses to ON stimulations ( n = 60 synaptic responses; n = 6 neurons; n = 6 larval animals; amplitude increase of 174 . 4 ± 167 . 0% , p < 0 . 05; no statistical differences between control and washout; Fig 4A ) . However , stimulation of the ON does not activate MOB neurons specifically , as it also activates medOB neurons . To rule out any involvement of the medOB in the increased RS responses after MOB gabazine injections , the effect of an electrical stimulation of the MOB with a gabazine injection ( 0 . 1 mM , 2 . 9 ± 1 . 1 nL ) in the MOB was tested . Under control conditions , MOB stimulation did not induce responses in RS neurons . However , after a gabazine injection in the MOB , electrical stimulation of the MOB elicited responses in RS cells ( n = 70 synaptic responses; n = 7 neurons; n = 7 larval animals; amplitude increase of 286 . 3 ± 296 . 8%; p < 0 . 05; no statistical differences between control and washout; Fig 4B ) . As a further control , electrical stimulation of the MOB under gabazine elicited significant responses in RS cells , even when the medOB had been surgically resected ( n = 5 larval animals , S5 Fig ) . Furthermore , recordings of the ventral roots of the spinal cord showed that electrical stimulation of the MOB after a gabazine injection in the MOB can induce fictive swimming ( in 65 . 1% of trials; 54 locomotor bouts out of 83 trials for gabazine versus 0 out of 105 for control; n = 9 larval animals , Fig 5 and S1 Data ) . Taken together , these findings suggest the presence of a previously unknown pathway linking the MOB to RS cells that seems to be under a strong tonic GABAergic inhibitory control . We investigated the spatial organization of projections from the MOB that would eventually reach the RS neurons . We injected the axonal tracer biocytin in the MOB ( n = 13 adult animals , Fig 6A1 ) and found ipsilateral axonal projections to the lateral pallium ( LPal ) , medial pallium , dorsal pallium , striatum , dorsomedial telencephalic neuropil , and habenula . Contralateral projections were found to the OB , dorsomedial telencephalic neuropil , striatum , and LPal . The MOB injections did not label any fibers in the PT . Similar olfactory projections from the OB have been reported in other species of lampreys [46 , 47] , but the selective contribution from the medOB or the MOB was not investigated in these earlier studies . The LPal appears to be a major target of neurons in the MOB , judging by the numerous labeled fibers seen to enter this region . The fibers densely filled the outermost layer covering the entire rostro-caudal extent of the LPal ( Fig 6A2 ) . Many fibers were also seen in the more central layers of the LPal , where the neuronal cell bodies of that structure are located . Tracer injections in the LPal ( n = 9 adult animals , Fig 6B1 ) retrogradely labeled many neurons in the MOB without ever labeling cell bodies in the medOB ( Fig 6B2 ) . The retrolabeled neurons in the MOB were found close to the glomeruli , but were almost never seen inside them . Physiological experiments were then carried out to characterize the effect of the pharmacological inactivation of the LPal on the responses of RS cells to the electrical stimulation of the MOB . Based on the results reported in Fig 4B , these experiments were carried out after removing the local GABAergic inhibition with a gabazine microinjection into the MOB ( 0 . 1 mM , 0 . 9 ± 1 . 0 nL , just prior each stimulation ) . An injection of glutamate receptor antagonists ( 2-amino-5-phosphonopentanoic acid [AP5]: 0 . 5 mM , 6-cyano-7-nitroquinoxaline-2 , 3-dione [CNQX]: 1 mM , 5 . 2 ± 0 . 8 nL ) in the LPal strongly decreased the RS neuron responses ( amplitude decrease of 64 . 8 ± 21 . 3%; p < 0 . 05 ) , thus confirming the role of the LPal in relaying glutamatergic outputs from the MOB to locomotor control centers ( n = 50 synaptic responses; n = 5 neurons; n = 5 larval animals , Fig 6C ) . Biocytin was injected in the LPal to examine its descending projections . Emphasis was placed on regions known to be involved in the medial olfactomotor pathway , such as the PT and the MLR ( n = 5 adult animals , Fig 7 ) . Numerous fibers terminated in the PT , predominantly on the ipsilateral side ( Fig 7B ) , with fibers crossing locally to the contralateral side ( arrows in Fig 7B2 ) . At levels immediately caudal to the PT , in the rostral mesencephalon , the number of descending fibers decreased sharply . Only a few labeled fibers continued to the level of the MLR ( Fig 7C ) , where many appeared to terminate ( Fig 7C2 ) . More caudal levels were not investigated in the present study , but it is not excluded that some fibers continued down more caudally [48] . Tracing experiments were carried out to further characterize the population of LPal neurons projecting to the PT and the MLR . The organization and anatomical boundaries of the LPal in lamprey are still debated [47 , 49–56] . In the present study , we followed the nomenclature of Northcutt and Puzdrowski [47] and Pombal and Puelles [54] . The part of the brain that was considered to be the LPal in the present study is illustrated in S6 Fig . In this series of experiments , a solution containing Texas Red-conjugated dextran amine ( TRDA ) was injected in the MOB to label olfactory projections from the OB and a solution containing biocytin was injected in the PT ( n = 7 adult animals and 1 larval animal , Fig 8A and S7 Fig ) or the MLR ( n = 4 adult animals and 1 larval animal , Fig 9A and S7 Fig ) to retrogradely label neurons projecting to the PT or MLR . Typical results are shown in Figs 8 and 9 and S7 Fig . Labeled cell bodies were distributed uniformly in all regions of the LPal , dorsal , ventral , rostral and caudal , when injections were made in the PT ( Fig 8A and 8B ) or the MLR ( Fig 9A and 9B ) . The dendrites of cells often extended radially towards the outermost layer of the LPal , where secondary olfactory fibers , labeled from the MOB , are located ( Fig 8B and Fig 9B ) . These results show that fibers originating in the MOB came in proximity with LPal neurons projecting to both the PT and the MLR , suggesting that the LPal is a relay for MOB inputs to the PT and MLR . Electrophysiological experiments were then conducted to examine the effect of deactivating the PT and the MLR on the RS neuron responses to the electrical stimulation of the LPal ( Fig 8C and Fig 9C , respectively ) . Glutamate antagonists were locally injected in either the PT ( AP5: 0 . 5 mM , CNQX: 1 mM , 1 . 1 ± 1 . 2 nL , Fig 8C ) or the MLR ( AP5: 0 . 5 mM , CNQX: 1 mM , 3 . 6 ± 2 . 5 nL , Fig 9C ) , and the RS neuron responses were markedly decreased ( PT: amplitude decrease of 48 . 7 ± 19 . 7%; p < 0 . 05; no statistical differences between control and washout; n = 50 synaptic responses; n = 5 neurons; n = 5 larval animals; MLR: amplitude decrease of 45 . 3 ± 21 . 7%; p < 0 . 05; n = 60 synaptic responses; n = 6 neurons; n = 6 larval animals ) . Taken together with our previous findings , these results show that glutamatergic olfactory outputs from the MOB are relayed via the LPal to the PT and to the MLR before reaching RS cells . The relative importance of the projection from the LPal to the PT or to the MLR was examined by counting retrogradely labeled cells in the LPal after an injection of a fluorescent tracer in the PT or the MLR . Bilateral biocytin injections in the PT ( n = 6 adult animals ) followed by the analysis of 10 LPals revealed that , on average , 751 ± 283 LPal neurons ( per LPal ) projected to the PT ( Fig 10A ) . Bilateral biocytin injections in the MLR ( n = 5 adult animals ) followed by the analysis of eight LPals revealed an average of 93 ± 62 neurons ( per LPal ) in these animals ( Fig 10A ) . The size of LPal neurons projecting to the PT and MLR was measured along their long axis . LPal PT- and MLR-projecting neurons measured on average 16 . 3 ± 3 . 0 μm ( n = 90 cells from a subset of three animals , Fig 10B ) and 15 . 4 ± 2 . 4 μm ( n = 90 cells from a subset of three animals , Fig 10B ) , respectively . Interestingly , a few medOB neurons were systematically labeled after an MLR tracer injection ( Fig 9A ) , thus demonstrating a direct projection from the medOB to the MLR . The lateral olfactomotor pathway ( orange pathway in Fig 11 ) may contribute significantly to the motor responses of lampreys to olfactory cues in their environment , in parallel to the previously described medial olfactomotor pathway ( green pathway in Fig 11 ) .
Lampreys , like many other animal species , display sex- and life stage–specific olfactory-induced motor behaviors [60–63] . The neural mechanisms accounting for the behavioral variability associated with a specific neural pathway within a species are largely unknown . However , the long-standing hypothesis that it was due to fundamental differences in brain wiring is now being challenged ( reviewed in [64] ) . Indeed , only very subtle sex-specific differences have been found in the structure and circuitry of the brain in mammals [65–68] . Likewise , we have shown in a previous study that a hardwired olfactomotor pathway is present in both sexes at all life stages in the sea lamprey [12] . For this reason , we hypothesized in the present study that modulatory mechanisms acting on this pathway could play a role in the variability of the behavioral responses of lampreys to olfactory cues . The OB is the first relay of the olfactomotor pathway . As such , it interfaces sensory afferents with motor control centers and it is ideally located to modulate olfactory-induced motor responses in lampreys . It has been proposed that the main function of the OB in vertebrates is the filtering and transmission of olfactory inputs [69] . Studies in mammals and turtles have shown that the sensory inputs to the OB are modulated both at presynaptic and postsynaptic levels by two classes of local GABAergic interneurons: periglomerular and granule cells [69] . Periglomerular cells inhibit glutamate release from primary olfactory axon terminals via a GABAB-mediated mechanism [70–73] . On the other hand , granule cells inhibit projection neurons via a GABAA-mediated mechanism [74–78] . Despite a rather good understanding of the cellular mechanisms responsible for the modulation of olfactory inputs , little is known about their overall effect on the OB output and , ultimately , on the resulting behavior . Using an in vitro isolated preparation of lamprey CNS , we provide the first evidence linking cellular GABAergic modulatory mechanisms in the OB to the activation of a sensorimotor pathway producing locomotor behavior . We showed that the lamprey OB anatomical organization is very similar to that of other vertebrates , regarding its GABAergic circuitry . Our material confirms previous work showing that the lamprey OB contains numerous GABAergic neurons of different morphological types [43] . The morphology and location of the GABAergic neurons suggest that they are mainly granule cells [43 , 45] , but not excluding possible periglomerular cells [43 , 51] . We also showed that both medOB and MOB glomeruli are densely innervated with GABAergic processes . These GABAergic processes are in close proximity to both primary olfactory axon terminals and dendrites or somata of OB projection neurons; this suggests possible pre- or postsynaptic contacts ( S2 Fig ) . We have not formally identified types ( i . e . , axons versus dendrites ) and origin ( i . e . , intrinsic versus extrinsic ) of the GABAergic processes . The abundant GABAergic cell bodies labeled in the OB suggest that they may be of intrinsic ( OB ) origin ( i . e . , granule cells or periglomerular cells ) , as seen in other vertebrate species [39–43] . The lamprey granule cells are axonless [45] , as in other vertebrate species . These processes are thus likely to be dendrites of granule cells or dendrites and axons of periglomerular cells , but some of these processes could be axons originating from neurons located in other parts of the brain . In mammals , most neuromodulatory inputs to the OB originate from the locus coeruleus ( noradrenergic inputs ) , the nucleus of the diagonal band of Broca ( cholinergic inputs ) , and the midbrain raphe ( serotoninergic ) ( reviewed in [69 , 79 , 80] ) . However , some cells located in the nucleus of the diagonal band of Broca are GABAergic and project to the OB [81 , 82] . We now show that injection of the GABAA receptor antagonist , gabazine , in the OB potentiates RS cell responses to ON or OB stimulation , thus suggesting an enhancement of the olfactomotor transmission . In the case of OB ( MOB ) stimulation , however , we cannot completely exclude that electrical stimulation of the MOB might recruit not only projection neurons but also local GABAergic interneurons , and that in such a case , an injection of gabazine might block the effect of their activation . Under gabazine , the electrical stimulation of the ON or OB can induce fictive swimming—the in vitro corollary of swimming behavior . Overall , these findings suggest that the GABAA antagonist gabazine increases the output of the OB . Indeed , downstream relays of the olfactomotor pathway ( PT and MLR ) control locomotion in a graded fashion [12 , 15 , 83] . Consequently , the increased RS cell responses observed under gabazine are likely to result from an increased drive from the OB to the PT and/or MLR . Studies in mammals have shown that GABA acts at several locations in the OB . OSN terminals express GABAB receptors [84–86] , which inhibit transmission from OSN axons to mitral cell primary dendrites upon release of GABA by periglomerular cells [70 , 71 , 87 , 88] . Mitral cell dendrites express both GABAA and GABAB receptors [89–93] . Pharmacological blockade or genetic alteration of GABAA receptors in mitral cells alters the OB γ oscillations and leads to increased ON-induced mitral cell discharges [78 , 94] . The effect of GABAB receptor activation in these cells is less clear [93] . Both periglomerular and granule cells release GABA on mitral cells; periglomerular cells contact mitral cell primary dendrites , whereas granule cells contact mitral cell secondary dendrites [69 , 95] . Granule and periglomerular cells also express GABAA receptors [89 , 91 , 96 , 97] . Genetic alteration of the GABAA receptor subtype expressed in granule cells ( i . e . , expressing the β3 subunit ) either globally or in a cell-specific manner increases the granule cell inhibition of mitral cells and results in increased OB γ oscillations [98 , 99] . To the best of our knowledge , the effect of periglomerular cell GABAA receptor activation on mitral cell activity has not been investigated , but an inhibition of periglomerular cells leading to the disinhibition of mitral cells could be expected . Finally , electrophysiological evidence suggests that granule cells also possess GABAB receptors whose activation modulates granule cell inhibition of mitral cells [100] . GABA can thus depress or potentiate mitral cell activity depending on its site of action ( i . e . , OSNs axons , OB interneurons , or mitral cell ) . However , as OB interneurons act on mitral cells via GABAA receptors , the net effect of the pharmacological blockade of GABAA receptors in all OB layers is likely to be a disinhibition of mitral cells . This is consistent with our results in lampreys and those found in other vertebrate species [77 , 94 , 101–104] . The presence of both tonic and phasic inhibition in the OB has been reported in fish , amphibians , and mammals [71 , 77 , 78 , 94 , 98 , 99 , 102 , 105 , 106] . It has been suggested that tonic inhibition may modulate the strength of sensory inputs to the OB [88] or the sensitivity of second-order olfactory neurons to sensory inputs [102] . Phasic inhibition has been shown to generate neuronal synchrony ( i . e . , oscillations ) in projection neurons [98 , 99] . The role of these oscillations and thus of the phasic inhibition is still debated , but several studies in insects and mammals point toward a crucial role in coding olfactory information [107–109] . Our study shows that a strong GABAergic inhibition of the OB output is present in the lamprey , one of the most basal extant vertebrate , and thus may be a common ancestral feature of the vertebrate OB . The GABAergic modulation of the olfactomotor pathways seen in lampreys could explain some of the life stage–specific behavioral responses to olfactory cues . For instance , migratory pheromones attract only pre-spawning adult lampreys [24–26] . This is surprising because these pheromones evoke strong responses in OSNs at other life stages [110] . Somehow , the activation of OSNs only leads to locomotor responses during the pre-spawning adult life stage . Meléndez-Ferro and colleagues [43 , 111] have stated that the density of OB GABAergic cells declines significantly between the newly transformed and pre-spawning life stages . Whether this apparent decrease in GABAergic cell density could account for some of the life stage differences is not known at present , but it could be one plausible mechanism worth investigating . A series of recent studies have shown that a CO2-mediated water acidification significantly impairs several olfactory-driven behaviors in fish , including prey tracking , predator avoidance , alarm response , and homing [112–116] . The mechanism at play has not been fully characterized yet , but it involves an alteration of the normal functioning of GABAA receptors , as blocking these receptors with gabazine led to a behavioral recovery [115 , 117] . The authors of these studies proposed that a potentiation or a reversal of the GABAA receptor function ( from inhibitory to excitatory ) because of changes in anionic gradients over neuronal membranes could underlie these behavioral alterations . Taken together , these studies show that GABAergic mechanisms also play a crucial role in modulating olfactomotor behaviors in fish . Further studies are needed to establish whether the neural pathways and modulatory mechanisms characterized in lampreys are also present in fish and other vertebrates . In the present study , we showed that stimulation of the MOB under gabazine led to excitatory responses in RS cells and to locomotion . This suggests the existence of a distinct pathway from the MOB to the RS cells and the presence of a strong tonic GABAergic inhibition in the MOB . We characterized the anatomy and physiology of this pathway . Anatomical data showed that the LPal receives a massive projection from the MOB and projects down to both the PT and MLR . The PT , in turn , projects to the MLR [12 , 118 , 119] . We also showed that the MLR receives a direct projection from the medOB , in addition to the already characterized projection via the PT [12] . The MLR then reaches the command cells for locomotion , the RS cells , via glutamatergic and cholinergic projections [120–123] . Physiological data confirmed that the LPal relays MOB olfactory inputs to the RS cells via the PT and MLR . This is consistent with the recent findings of Suryaranayana and colleagues [124] indicating that some LPal neurons receive monosynaptic inputs from the OB . Interestingly , Ocaña and colleagues [48] showed that a few fibers originating in the LPal could reach RS neurons directly and that some of these could be followed as far as the first spinal segments . This prompted the authors to conclude that the LPal possesses an efferent projection pattern similar to that of the amniote motor cortex [48] . It would be interesting to examine if these projections from the LPal to the RS neurons and spinal cord are also involved in olfactomotor responses . Although we cannot exclude that there may be other pathways linking olfactory centers to motor centers , our study demonstrates the existence of two distinct glutamatergic pathways linking the olfactory and motor systems in lampreys ( Fig 11 ) . Both these pathways share a common output via the PT/MLR–RS neurons system . However , they differ regarding their pathways from the OB to motor control centers ( i . e . , PT/MLR ) , as well as to their inputs from the periphery . In lampreys , the main olfactory epithelium contains numerous tall , ciliated OSNs expressing the G-protein Golf , as in the main olfactory epithelium in other vertebrates [125–133] . The OSNs of the main olfactory epithelium project their axons to the MOB , which , in turn , projects mainly to the LPal , i . e . , the putative homologue of the mammalian olfactory cortex in lampreys [134] . This pathway is strikingly similar to the main olfactory pathway of terrestrial vertebrates and thus further supports its evolutionary conservation . In addition to the main olfactory epithelium , lampreys possess an accessory olfactory organ [29–32 , 135–138] . The accessory olfactory organ contains short , broad , ciliated OSNs [32] that do not express the G-protein Golf and project only to the medOB [32 , 129] . Projection neurons of the medOB then project directly to the PT and MLR , bypassing the LPal . Taken together , these findings show that the lamprey accessory olfactory organ constitutes a discrete olfactory subsystem . It has even been suggested that the accessory olfactory organ represents a primordial vomeronasal system [29 , 31 , 138] . In other vertebrates , the presence of parallel olfactory pathways conveying the information from the periphery to high-order brain olfactory centers suggests that these systems subserve different behavioral functions [139–143] . For instance , in fish , segregated olfactory pathways , from the olfactory epithelium to the telencephalon , mediate feeding , reproductive , and alarm behaviors [139 , 142 , 144–151] . Similarly , the main and accessory ( i . e . , vomeronasal ) systems of terrestrial vertebrates are segregated until at least the third-order neurons and their respective activation elicits different behaviors [152–155] . Physiological evidence in lampreys also supports this hypothesis , as OB local field recordings showed that the medOB and MOB have overlapping but different response profiles to feeding cues and pheromones [33 , 34] . Moreover , we show that the pathways from the OB to the motor control centers differ for the two olfactory subsystems . The medOB projects directly to motor control centers , whereas the MOB projects first to the LPal before reaching motor control centers . Not surprisingly , activation of both systems leads to locomotion . This could be attributed to the paucity of the behavioral repertoire of lampreys compared to mammals . However , it should be noted that reproductive , migratory , and feeding behaviors all require locomotion in lampreys . The distinction between these two subsystems thus lies in their inputs from the periphery ( accessory olfactory organ versus main olfactory epithelium ) as well as in the involvement of the LPal in the lateral pathway . It is tempting to propose that the medial pathway could mediate innate responses to chemical stimuli ( for example , avoidance ) , whereas the lateral pathway could be involved in olfactomotor behaviors requiring further processing and perhaps learning ( for example , olfactory navigation ) . A similar distinction between dual “olfactory” systems exists in invertebrates [156–158] . In mammals , it was shown that mitral cells of the MOB can develop differential responses to rewarded/unrewarded odors [159] . It has been suggested that the dichotomy between innate responses versus learned responses may be what distinguish the main and accessory systems of terrestrial vertebrates [154] . This hypothesis has , however , received little attention , and further studies are needed . In conclusion , our study shows that olfactory inputs can activate the locomotor command system via two distinct glutamatergic pathways in lampreys . To the best of our knowledge , this is the first characterization of a dual olfactory pathway , from the periphery to the motor command system , in vertebrates . Both pathways are strongly modulated by the GABAergic circuitry of the OB that may account for some of the variability in behavioral responses to olfactory inputs in lampreys . The existence of two segregated olfactory subsystems in one of the most basal extant vertebrates sheds light on the evolution of the olfactory system and suggests that its organization in functional clusters could constitute a common ancestral trait of vertebrates .
For all procedures , the animals were deeply anesthetized with tricaine methanesulphonate ( MS-222 , 200 mg/L , Sigma-Aldrich , Oakville , ON ) and then decapitated . All surgical and experimental procedures conformed to the guidelines of the Canadian Council on Animal Care and were approved by the animal care and use committee of the Université de Montréal ( Protocol no . 18–018 ) , the Université du Québec à Montréal , and the University of Windsor . Experiments were performed on 57 larval and 61 adult sea lampreys ( Petromyzon marinus ) of both sexes . Some animals were used in more than one experiment . Larvae were collected from the Pike River stream ( QC , Canada ) . Adults were collected from the Great Chazy River ( NY , United States ) and were kindly provided by agents of the U . S . Fish and Wildlife Service of Vermont . The permission to collect animals in the field was granted by the Quebec's Ministry of Natural Resources and Wildlife ( permit no . 2017-03-30-2189-16-SP ) . All animals were kept in aerated fresh water maintained at 4–5 °C . For all types of experiments , the animals were deeply anesthetized with tricaine methanesulphonate ( MS-222 , 200 mg/L , Sigma-Aldrich ) , decapitated caudal to the seventh branchiopore , and transferred into cold oxygenated Ringer's ( 8–10 °C ) of the following composition ( in mM ) : 130 NaCl , 2 . 1 KCl , 2 . 6 CaCl2 , 1 . 8 MgCl2 , 4 . 0 HEPES , 4 . 0 dextrose , and 1 . 0 NaHCO3 , at pH 7 . 4 . The branchial apparatus , myotomal musculature , and all soft tissues attached to the ventral side of the cranium were removed . The dorsal part of the vertebrae and cranium were removed to expose the brain and the rostral spinal cord . The peripheral olfactory organ was left intact with the ON still attached to the brain . All other nerves were cut and the choroid plexus covering the fourth and the mesencephalic ventricles was removed . The preparation was pinned down to the bottom of a recording chamber lined with Sylgard ( Dow Corning , Midland , MI ) and continuously perfused with cold oxygenated Ringer's ( about 4 mL/min ) . Intracellular recordings of RS neurons were performed under visual guidance through a M3C stereomicroscope ( Wild-Heerbrugg , Heerbrugg , Switzerland ) using sharp glass microelectrodes ( 60–130 MΩ ) filled with 4 M potassium acetate . The signals were amplified with an Axoclamp 2A ( 20 kHz sampling rate , Axon Instruments , Foster City , CA ) and acquired through a Digidata 1322A interface running on pClamp 9 . 2 software ( Axon Instruments , Foster City , CA ) . Only RS neurons displaying a stable membrane potential lower than −70 mV for at least 15 min were considered in this study . Electrical stimulation ( 1–3 pulses , 5–30 μA , 2-ms duration , and 20-ms pulse interval ) was delivered using homemade glass-coated tungsten electrodes ( 0 . 8–2 MΩ , 10–50 μm tip exposure ) connected to a Grass S88 stimulator via a Grass PSIU6 photoelectric isolation unit ( Astro-Med , Longueuil , QC ) . A delay of 50 s was allowed between each stimulation . In the figures , the synaptic responses are illustrated as the mean of 10 responses obtained with the same stimulation parameters . In some experiments , the left and right ventral roots from one spinal segment ( usually around the 10th segment ) were also recorded using extracellular glass electrodes ( tip diameter about 5 μm ) filled with the Ringer's solution . The signals were amplified ( ×10 , 000 ) and filtered ( 100 Hz–1 kHz band-pass ) using an AM systems 1800 dual channel amplifier ( AM systems , Sequim , WA ) and monitored for the presence of neural activity . “Fictive locomotion” ( originally defined by Perret and colleagues , 1972 ) [160] observed in the absence of muscles and movement was defined as a neural activation of the spinal ventral ( motor ) roots , with a similar pattern as myotomal contractions seen during swimming . Drugs were purchased from Sigma-Aldrich ( AP5 ) , Tocris Bioscience ( gabazine and CNQX ) , and Thermo Fisher Scientific ( Fast Green ) . They were kept as frozen concentrated stock solutions and dissolved to their final concentrations in Ringer's solution prior to their use . Gabazine ( SR-95531; 0 . 1–1 mM ) and the CNQX/AP5 mixture ( 0 . 5 mM/1 mM ) were pressure ejected ( 3–180 pulses; mean ± SD = 57 ± 46 pulses; about 4 psi , 20–40-ms pulse duration , injection volume: 0 . 1–6 . 6 nL; mean ± SD = 2 . 6 ± 2 . 1 nL ) through glass micropipettes ( 10–20 μm tip diameter ) in the brain tissue , using a Picospritzer ( General Valve Corp , Fairfield , NJ ) . The inert dye Fast Green was added to the drug solution to monitor the extent of the injections . The spread did not exceed 300 μm in diameter for any microinjection . RS cells were retrogradely labeled by placing crystals of the calcium-sensitive indicator dye Calcium-Green dextran ( 3000 MW , Invitrogen , Eugene , OR ) on the rostral stump of the spinal cord , transected at the level of the first spinal segments . The preparation was then kept in cold , oxygenated Ringer's solution for 24–36 h of axonal transport . Labeled cells were observed under a Nikon epifluorescence microscope equipped with a 20× ( 0 . 75 NA ) objective . A filter set appropriate for fluorescein isothiocyanate ( FITC ) was used to visualize the neurons . The emitted light was captured with an intensified CCD video camera ( Photometrics CoolSNAP HQ , Roper Scientific , Tucson , AZ ) and recorded at a rate of two images per second , using Metafluor imaging software ( Molecular Devices , Sunnyvale , CA ) . Calcium responses are expressed as relative changes in fluorescence ( ΔF/F ) . Results are presented as mean ± SD . Statistical analyses were performed using Sigma Plot 11 . 0 ( Systat , San Jose , CA ) , and statistical significance was set at p < 0 . 05 . To test for differences in mean between groups , we performed a one-way analysis of variance for repeated measures , followed by a Holm-Sidak’s multiple-comparison post hoc test or Friedman analysis of variance for repeated measures on ranks followed by Tukey multiple-comparison post hoc test . | Olfactory-induced behaviors ( homing , food or mate searching , etc . ) are crucial for the survival and reproduction of most animals . A neural substrate underlying odor-induced behaviors in vertebrates was recently uncovered using a basal vertebrate model: the lamprey . It consists of a neural pathway extending from the medial olfactory bulb , a first-order relay of olfactory information in the brain , to locomotor regions . Here , we investigated modulatory mechanisms acting on this neural pathway . We show that an inhibitory circuitry that releases the neurotransmitter GABA in the olfactory bulb strongly modulates motor responses to olfactory stimulation . We also discovered and characterized a novel olfactomotor pathway that originates in the non-medial olfactory bulb and consists of a projection to the lamprey olfactory cortex that , in turn , projects to locomotor regions . This discovery of a novel pathway linking olfactory and motor centers in the brain indicates that olfactory inputs can activate locomotor centers via two distinct pathways . Both pathways are strongly modulated by the neurotransmitter GABA in the olfactory bulb . The existence of segregated olfactory subsystems in lampreys sheds light on the evolution of olfactory systems in vertebrates . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"fish",
"brain",
"vertebrates",
"neuroscience",
"animals",
"biological",
"locomotion",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"lampreys",
"olfactory",
"organs",
"functional",
"electrical",
"stimulation",
"animal",
... | 2018 | GABAergic modulation of olfactomotor transmission in lampreys |
If perturbing two genes together has a stronger or weaker effect than expected , they are said to genetically interact . Genetic interactions are important because they help map gene function , and functionally related genes have similar genetic interaction patterns . Mapping quantitative ( positive and negative ) genetic interactions on a global scale has recently become possible . This data clearly shows groups of genes connected by predominantly positive or negative interactions , termed monochromatic groups . These groups often correspond to functional modules , like biological processes or complexes , or connections between modules . However it is not yet known how these patterns globally relate to known functional modules . Here we systematically study the monochromatic nature of known biological processes using the largest quantitative genetic interaction data set available , which includes fitness measurements for ∼5 . 4 million gene pairs in the yeast Saccharomyces cerevisiae . We find that only 10% of biological processes , as defined by Gene Ontology annotations , and less than 1% of inter-process connections are monochromatic . Further , we show that protein complexes are responsible for a surprisingly large fraction of these patterns . This suggests that complexes play a central role in shaping the monochromatic landscape of biological processes . Altogether this work shows that both positive and negative monochromatic patterns are found in known biological processes and in their connections and that protein complexes play an important role in these patterns . The monochromatic processes , complexes and connections we find chart a hierarchical and modular map of sensitive and redundant biological systems in the yeast cell that will be useful for gene function prediction and comparison across phenotypes and organisms . Furthermore the analysis methods we develop are applicable to other species for which genetic interactions will progressively become more available .
One of the major goals in biology is to understand how molecules are organized within the cell , how they interact to mediate biological processes and how process failure leads to disease . Genetic perturbations , such as gene mutations , are often used to better understand the function of a gene and to study the relationship between genotype and phenotype [1] . In budding yeast , most genes ( ∼80% ) are not essential for growth under standard laboratory conditions [1] , suggesting that their function is not required under the conditions tested or is compensated by other genes . Exploring mutant phenotypes in the presence of a chemical or an environmental stress [2] , along with combining multiple mutations to map genetic interactions [3] , [4] , [5] , have been successful strategies to investigate genetic redundancy . In particular , genetic interactions have proven useful to predict gene function [6] and organize biological processes [7] , [8] , [9] and are complementary to other functional interaction data such as protein-protein interactions [10] . Here we systematically evaluate how genetic interaction data relates to known biological processes . We next review previous work in this area to place our work into context . Genetic interactions are observed when the phenotype of a double mutant is unexpected given the phenotypes of both single mutants [11] . With respect to growth , a genetic interaction is classified as either positive ( or negative ) when the fitness of the double mutant is higher ( or lower ) than expected . Negative genetic interactions often indicate functional redundancy between two genes , with the extreme case being synthetic lethality ( SL ) when simultaneous deletion of two otherwise non-essential genes leads to cell death . A biochemical interpretation for this is that the two genes participate in complementary or parallel pathways or complexes [12] , [13] . As a result , two complementary pathways tend to be connected by many negative genetic interactions . Positive interactions may indicate a number of biochemical scenarios , but are typically thought of as being within a pathway or complex , such as a linear chain of reactions where the deletion of one gene affects output , but deleting a second gene doesn't further affect output [4] . While the precise relationship of a genetic interaction to its underlying biochemistry is still not completely understood , functionally related genes tend to have similar genetic interaction profiles . Clustering the first large scale genetic interaction maps composed of SL and synthetic sick interactions resulted in clear grouping of genes with correlated genetic interaction profiles [4] , [5] , [6] , [7] . These genes tend to be physically linked as part of the same biochemical pathway , multiprotein complex or physically interacting protein pairs . Genes with correlated profiles tend to encode physically interacting proteins and tend not to interact synthetically with each other [4] , [6] . This supports a model where parallel biochemical pathways are connected by a large number of SL interactions [14] , [15] , [16] . Extending the parallel pathway model , Kelley et al . defined biological modules as clusters of proteins enriched in genetic and physical interactions ( “within-module” ) connected only by SL genetic interactions ( “between-module” ) [17] . This approach was further explored by defining modules more generally as a connected graph in a protein interaction network [18] and considering pairs of modules connected by negative interactions [19] . Extending this idea to quantitative positive and negative genetic interactions , Bandyopadhyay et al . and others devised methods to learn protein complexes and their functional relationships [20] , [21] . These approaches aim to define functional modules ( clusters of functionally related genes , such as pathways or complexes ) using large-scale genetic interactions . By predicting positive and negative interactions on a large scale from metabolic network simulations , Segre et al . discovered that biological modules are often connected by purely positive or negative genetic interactions , and described these as monochromatically pure connections [22] . Quantitative experimentally determined genetic interactions [4] , [7] have been interpreted using this concept and found to also contain monochromatic groups of genes , where all genes are connected to each other by predominantly positive or negative genetic interactions . Theoretically , we can consider four monochromatic patterns: monochromatic positive or negative interactions within groups of genes ( “within module” ) and monochromatic positive or negative connections between groups of genes ( “between module” ) . Only some of these patterns have been related to the underlying biochemical system or systematically explored . Monochromatic positive within module patterns have been shown to correspond to protein complexes or pathways [4] , [9] , [20] , [23] , [24] . Monochromatic negative within module patterns have been observed to represent complexes containing essential genes [20] , [25] . Also , complexes enriched in genetic interactions tend to be monochromatic positive or negative [25] . Monochromatic positive between module connections have not been related to a biochemical model , but have been observed to occur between functional modules in simulations [22] and between complexes [25] . Monochromatic negative between module connections have been observed in simulations [22] and are expected from the observation that SL interactions ( negative ) connect parallel pathways [9] , [14] , [15] . Overall , monochromatic patterns have been linked to various physical modules , however none have been systematically studied in terms of all known pathways and complexes . Two approaches to systematically study monochromatic patterns are possible . We can either search for monochromatic patterns in the genetic interaction network and interpret the results in terms of known physical and functional modules ( such as protein complexes or biological processes ) , or we can examine all known modules for monochromatic patterns . The former approach has been applied in a non-exhaustive fashion on focused genetic interaction data sets [26] , [27] , but an exhaustive approach is computationally difficult , given the large size of recently published genetic interaction networks [7] , and requires the development of new algorithms . The latter approach uses the knowledge we currently have about functional modules to study monochromatic patterns and this is what we adopt here . We use current knowledge about biological processes from Gene Ontology annotation [28] combined with the most comprehensive quantitative genetic interaction data set currently available , which includes measurements for 5 . 4 million gene pairs in normal growth conditions and provides quantitative genetic interaction profiles for ∼75% of all genes in Saccharomyces cerevisiae [7] . We first assess the monochromatic nature of biological processes and their connections and find that only 10% of biological processes and less than 1% of inter-process connections are monochromatic . We next explore various features that may explain these monochromatic patterns and show that protein complexes are responsible for a surprisingly large fraction of them . Significantly more genetic interactions than expected are attributed to complexes and genes encoding protein complex members have more genetic interactions and are essential more often than expected . This work shows the importance of protein complexes in contributing to monochromatic patterns in quantitative genetic interaction networks and generates a hierarchical and modular map of sensitive and redundant biological systems in the yeast cell .
To study the monochromatic nature of known biological processes , we used the most recent data set of quantitative genetic interactions , generated by Synthetic Genetic Array ( SGA ) analysis [7] . Known processes were defined by the Gene Ontology ( GO ) biological process ( BP ) classification system as annotated to yeast genes by the Saccharomyces Genome Database ( SGD ) . GO annotations in yeast are the most complete for any organism and there is no other comparable database of biological processes for yeast . Processes can represent canonical pathways , like fatty acid biosynthesis , and also more general processes , like DNA repair . Each process is described by a standard name and a set of genes annotated to it . We considered all GO processes in yeast where member genes were connected by at least one SGA interaction ( ∼1000 processes ) . We defined the monochromatic score as the relative ratio of positive to negative interactions occurring within a given process ( set of genes ) . To assess how likely these scores are to occur by chance , we computed Z-scores using randomization that maintains the network topology ( permutation of the gene names ) . We can then identify unexpected monochromatic patterns by their high Z-scores ( Figure 1 ) . Highly positive Z-scores characterize monochromatic positive processes and highly negative Z-scores characterize monochromatic negative processes ( Methods ) . Not all gene pairs are tested in the SGA data set , thus processes have variable genetic interaction coverage . Because monochromatic patterns are more confident for processes that have an increased coverage of genetic interactions , we selected high coverage processes based on the number of corresponding genes present and connected in the SGA genetic interaction network ( Methods ) . For a given coverage level , we computed the ratio of monochromatic processes among all covered processes . We found that this ratio ranges from 7 to 9% ( Table 1 ) . Thus , just under 10% of SGA covered biological processes are monochromatic . Choosing a coverage cut-off of 0 . 6 that reasonably traded higher coverage for a larger number of terms , we identified 50 monochromatic processes , including 5 positive and 45 negative ( Dataset S1 ) . Thus , even though positive interactions are often presumed to be acting within processes [4] , [9] , there are actually more processes that are monochromatic negative . Monochromatic processes are functionally diverse , but also biased . For instance , microautophagy and histone exchange are monochromatic positive whereas protein import and small GTPase mediated signal transduction are monochromatic negative ( Figure 2 ) . Globally , monochromatic processes are enriched in specific functions , including chromosome segregation/microtubule and protein degradation/proteasome ( Figure 3 ) . Further , positive monochromatic processes are generally much smaller ( < = 40 genes ) than negative ones ( ∼100 genes ) and are more specific in the GO hierarchy ( Figure S1 ) . To investigate the monochromatic nature of connections between biological processes , we defined an inter-process connection as the set of genetic interactions linking genes annotated to two processes , excluding those annotated to both . The monochromatic nature of an inter-process connection was assessed by measuring enrichment for positive and/or negative interactions relative to the background distribution in the entire genetic interaction network ( Methods ) . The resulting p-value was used as a score to select the most monochromatic connections . The coverage of a connection was assessed by the number of genetic interactions tested in SGA compared to all possible interactions between the two processes . Using a range of coverage cut-offs , we found that only ∼0 . 27% of covered connections are monochromatic ( Table 2 ) . For instance , the pair of processes ‘glycosylation’ ( 49 genes ) and ‘tRNA modification’ ( 51 genes ) is connected by 44 positive genetic interactions and 29 negative interactions while we expect 20 positive and 36 negative interactions by chance . This connection is thus highly enriched in positive interactions . The process ‘tRNA modification’ is connected to ‘cell wall organization’ ( 191 genes ) by 80 positive genetic interactions and 296 negative interactions while we expect 76 positive and 137 negative interactions by chance . This predominantly negative connection indicates that many genes from both processes buffer each other . With a reasonable coverage cut-off of 0 . 6 , we identified 1 , 387 monochromatic connections , including 614 positive ( 44% ) and 773 negative connections ( 56% ) ( Dataset S2 ) . Previous analyses suggested that positive interactions often occur between genes acting together in the same pathway or complex [4] , [9] and negative genetic interactions tend to occur between genes implicated in redundant processes [9] , thus we expect connections to be monochromatic negative . However , we observe an even distribution of negative to positive connections showing that between process connections are not predominantly negative . In addition , our results show that connections between GO-defined processes are rarely monochromatic . We noticed that the monochromatic processes identified above often contain protein complexes or parts of complexes . In fact , all monochromatic processes but six contain at least one gene encoding a member of a complex . Since complexes enriched in genetic interactions tend to be monochromatic [25] , we evaluated the contribution of all protein complexes to the monochromatic patterns we observed . To do this , we removed genes or interactions corresponding to protein complexes from the SGA genetic interaction data set and repeated our monochromatic analysis described above . When we removed all genes encoding proteins that are part of a complex , most ( 82% ) of the monochromatic processes identified above were no longer monochromatic ( Figure 4 ) . Interestingly , a few new processes became monochromatic after this , but these were either small processes with positive interactions or very large processes with negative interactions ( Figure S1 ) . When we removed interactions occurring within complexes , a smaller number ( 28% ) of the monochromatic processes were explained . These results hold for various coverage cut-offs ( Table 1 in Text S1 ) . As a control , we showed that removing the same number of random genes encoding proteins not in any complex did not have the same effect ( Kolmogorov-Smirnov ( KS ) test p<4 10−4 , Figure S2 ) . We confirmed our results using another curated set of yeast protein complexes that combines predictions from high-throughput and literature data to form a consensus set [29] ( called Consensus ) . This data set includes existing complexes defined previously by Pu et al . and Hart et al . [30] , [31] . Again most monochromatic processes ( 90% ) were explained by genes encoding proteins in complexes ( Figure S3 ) and this result holds for various coverage cut-offs ( Table 3 in Text S1 ) . This indicates that genes whose products are part of a complex are the main contributors to the monochromatic genetic interaction patterns we see in GO biological processes . We also considered three other features that may contribute to our monochromatic patterns: essential genes , low single mutant fitness genes and duplicate genes . Essential genes ( tested as hypomorphic or conditional alleles in genetic screens ) are known to have many negative interactions [7] , genes which have a strong effect on yeast fitness , as measured by growth rate when deleted ( i . e . a low single mutant fitness ) similarly tend to show many negative interactions [7] , and duplicate genes often buffer each other and thus are typically connected by a strong negative interaction [32] . We removed each of these gene sets in turn and evaluated the effect on our observed monochromatic patterns . All these features partly explain the monochromatic patterns previously identified but not as much as the genes encoding proteins in complex ( Figure S3 ) . In addition , these features are highly overlapping with the set of genes encoding proteins in complexes ( Figure S4 ) . For example , 60% of essential genes are in a complex . Thus , we presume that the effect of these features on monochromatic processes is minor and mainly due to their correlation with protein complexes . To measure to what extent the features presented above explain the monochromatic nature of connections , we adopted the same strategy of removing each feature in turn and analyzing the resulting change in number of monochromatic connections . Again , genes encoding proteins in complexes explained most monochromatic connections ( 98% ) whereas the other features only partly explained the monochromatic connections ( Figure 4 , Figure S3 ) . This result holds for various coverage cut-offs ( Table 2 in Text S1 ) and was confirmed using Consensus , the alternative set of protein complexes ( Figure S3 , Table 4 in Text S1 ) . Removing the same number of random genes not in a complex did not have the same effect on the monochromatic pattern ( KS p = 0 , Figure S2 ) . An example of a monochromatic connection explained is the positive connection between the ‘mitotic sister chromatid cohesion’ and the ‘regulation of glucose metabolic process’ processes ( Figure S5 ) . This monochromatic positive connection is mainly due to positive genetic interactions between genes encoding proteins from the ‘GID complex’ and the ‘AMP-activated protein kinase complex’ on one side , and the ‘replication fork protection complex’ and the ‘Ctf18 RFC-like complex’ on the other side . When we removed the genes encoding proteins in a complex , these processes were no longer connected by a monochromatic positive connection . These results suggest that genes encoding proteins in a complex play a key role in the monochromatic connections between yeast GO biological processes . Since protein complexes are important in explaining monochromatic GO processes in the genetic interaction network , we examined their contribution at the genetic interaction level . We defined a genetic interaction as involving a complex if at least one gene in the interaction encodes a protein that is part of a complex . It is expected that 49% ( 93 , 383 ) of all observed SGA interactions ( 189 , 996 ) involve a protein complex gene since 49% ( 2 , 801 , 630 ) of all tested gene pairs involve a protein complex gene . Surprisingly , we found that 63% ( 119 , 871 ) of the observed SGA genetic interactions involve complexes , or 28% more than expected . This significant bias ( Fisher p<10−5 ) is present globally and for both negative and positive interactions ( Table 3 ) . Since some genes might be noisy and cause a false signal by virtue of having an extreme number of interactions , we repeated the analysis with progressively more stringent sets of genetic interactions , defined by the SGA score [25] . At each increased stringency level , we found the result to be stronger and more significant ( Table 3 in Text S1 ) . In addition , the global degree distribution confirms that genes encoding proteins in complexes are more likely to have more interacting partners than genes encoding proteins not in any complex ( KS p<2 10−4 ) . Finally , to check the robustness of our results to our definition of complexes , we confirmed them using the Consensus data set used above [29] ( Methods , Table 10 in Text S1 ) . These results indicate that genes encoding proteins in complexes are more likely to genetically interact than genes encoding proteins not in any complex . As previously noted , most monochromatic negative complexes contain essential genes [20] , [25] . More generally , we found that essential genes are highly enriched within complexes , 225% more than expected ( Fisher p = 0 ) ( Table 4 in Text S1 ) and this result also holds when considering only genes present in the genetic interaction network ( Fisher p<10−60 , Table 5 in Text S1 ) . Furthermore , the number of essential genes per complex has a broad distribution ( Figure S6 ) : many complexes are composed of all essential genes , and a high proportion ( 57% ) of protein complexes are essential ( contain at least one essential gene ) . This bias was also present in the Consensus data set ( Tables 6–9 in Text S1 ) . This result may be related to the increased number of genetic interactions involving protein complexes observed above , as essential genes are known to have more genetic interactions than non-essential genes [7] . This observation may be influenced by experimental preference for studying essential genes , but this can’t fully explain the results , as many high-throughput methods have been used to defined yeast complexes [29] . Altogether , the above results show that protein complexes play an important role in the monochromatic genetic landscape of biological processes and more generally in yeast growth .
Monochromatic patterns have been used to identify biological processes and other functional modules [22] , [26] , [27] . In this work , we ask to which extent known processes show these monochromatic patterns . To answer this question , we systematically studied the monochromatic landscape in yeast using known biological processes as defined by GO annotation and a large network of genetic interactions . We found that approximately 10% of GO-defined biological processes that are sufficiently covered by genetic interactions are monochromatic and less than 1% of all pairs of processes interact monochromatically . We observe that monochromatic processes tend to be predominantly negative whereas between process connections are evenly distributed between positive and negative . Interestingly , we found that protein complexes explain most of the monochromatic signal present in GO processes and are disproportionately important for yeast growth ( are involved in more genetic interactions and contain more essential genes compared to non-complex genes ) . We hypothesize that protein complexes are more sensitive to perturbation and more difficult to buffer , either because it is more difficult to duplicate the functionality of an entire complex or that complexes participate in more processes compared to individual proteins ( Figure S7 ) . Previous work observed that protein complexes are often monochromatic [7] , [20] but we show for the first time that the monochromatic patterns identified within and between biological processes are mainly driven by protein complexes . We chose GO as the representation of known biological processes since it is the most comprehensive resource available . KEGG and SGD YeastCyc also make available pathway information , but these are limited mostly to metabolic pathways and do not cover as many genes as GO , making a general analysis difficult . In addition , GO organizes processes hierarchically , which clarifies the relationships between processes and sub-processes . However , this makes processes highly overlapping . The number of monochromatic processes depends on this overlap . To assess the effect of overlap , we applied our method on the reduced ontology GO Slim , which contains fewer and less overlapping terms compared to the full GO . We identified 11 monochromatic processes among 26 covered processes ( Dataset S3 , Dataset S4 ) . Similarly to the full GO analysis , most ( 73% ) of the monochromatic processes are no longer monochromatic when removing genes encoding proteins complex members ( Figure S8 ) . In addition we applied the analysis to the MIPS Comprehensive Yeast Genome Database ( CYGD ) Functional Category ( FunCat ) , an alternative classification of biological processes [33] . This classification is hierarchical but , unlike GO , every category has only one root category . Since genes can be annotated to several categories , there is some overlap between categories , but to a lesser extent compared to GO . We found that 8% of the covered processes were monochromatic ( similar to GO ) ( Dataset S5 , Dataset S6 ) and protein complexes explained most ( 78% ) of them ( Figure S8 ) . This confirms that protein complexes play an important role in the monochromatic nature of biological processes , even when considering less overlapping process definitions . While we used the most comprehensive data for genetic interactions and process annotations available , much data is still missing which can impact our results . Some processes may appear monochromatic because not all interactions are known , or some processes may not be considered because they are lacking interactions . To account for the lack of completeness of the genetic interaction network , we only considered highly covered processes . The results are presented for a reasonable level of coverage and confirmed with multiple coverage thresholds . Also , monochromatic processes likely exist that contain genes that are not covered by our best efforts to collect the most comprehensive annotation available ( GO , FunCat , KEGG , YeastCyc ) . As annotation improves , we expect the monochromatic map to expand . Also , complementary unsupervised approaches , such as clustering or motif detection , can be used to find monochromatic patterns that we miss . It will be interesting to see how many monochromatic modules found by these methods are not currently captured in GO . The monochromatic processes , complexes and connections we find chart a hierarchical and modular map of sensitive and redundant biological systems in the yeast cell ( Dataset S1 , Dataset S2 , Figure S9 ) . Our results indicate that the genetic interaction network is enriched in interactions involving protein complexes , monochromatic connections between processes are rare and protein complexes play an important role in defining monochromatic patterns within and between processes . These results are illustrated on the example map presented in Figure S9 . Our map holds for genetic interactions measured in standard laboratory growth conditions . It will be interesting to compare it with other maps constructed based on genetic interactions defined using phenotypic readouts other than yeast growth or not in standard laboratory conditions [34] , [35] and from other species for which genetic interactions will progressively become more available , such as Caenorhabditis elegans [36] , [37] , Drosophila melanogaster [38] or mammalian cells .
The genetic interaction data are from the most recent and comprehensive study in yeast obtained by the Synthetic Genetic Array technique ( SGA ) in normal growth conditions [7] . This data set consists of 191 , 890 pair-wise interactions between 4 , 415 genes derived from 1 , 712 full genome screens . Each interaction is characterized by the epsilon score , a quantitative genetic interaction measure , and a p-value , indicating confidence . This score can be positive or negative , indicating a positive or a negative interaction . When different measurements are available for a single gene ( i . e . from several screened alleles of essential genes ) , we merge all interactions ( this occurs for 35 genes ) . If two screens give opposite scores , we remove both . If two screens give scores of the same sign , we keep the one with the best p-value . The resulting network contains 166 , 401 pair-wise interactions among 4 , 415 genes . We downloaded the annotation of the yeast genome provided by SGD [39] on September 7th , 2009 . For the Biological Process ontology , all genes annotated to one specific GO term are up-propagated to all parents of that GO term . We don't consider non-manually reviewed annotations ( IEA evidence code ) . We only consider GO terms with more than one observed interaction between its genes and with less than 200 genes in the genetic interaction matrix , otherwise the random networks are not different enough to assess the statistical significance of the monochromatic scores . We are left with a set of 1 , 031 processes in yeast with genetic interactions in SGA . In addition , we downloaded the functional categories from FunCat [33] and filtered out those not referring to biological processes ( 16: protein with binding function or cofactor requirement; 70: subcellular localization; 73: cell type localization; 75: tissue localization; 77: organ localization , 18: regulation of metabolism and protein function; 98: classification not yet clear-cut; 99: unclassified proteins ) . For a given GO term , its genes can be present in the genetic interaction network or not . If present , they contribute to the monochromatic nature only if they are connected by an SGA interaction within the GO term . We assess the coverage of the GO term by the minimum value of the two following ratios: ( i ) the number of genes in the GO term and in the SGA genetic interaction network over the total number of genes in the GO term; ( ii ) the number of connected genes in the GO term over the total number of genes in the GO term and in the genetic interaction network . We define the monochromatic score of a GO term as the relative ratio of positive to negative interactions observed between the genes in that GO term ( equation 1 ) . ( 1 ) where I is the set of interactions occurring between two genes from the GO term t . This score ranges from −1 , meaning fully monochromatic negative , to +1 , meaning fully monochromatic positive . We then generate random networks by shuffling the labels of the original genetic interaction network , which preserves the network topology . For each GO term , we compute a series of monochromatic scores obtained with the random genetic interaction networks and use this distribution of scores to compute a Z-score ( equation 2 ) . ( 2 ) where S is the monochromatic score to be standardized , μ is the mean of the random scores and σ the standard deviation of the random scores . GO terms with a Z-score larger than 1 . 6 in absolute value are selected as monochromatic . A connection between two GO terms is formed by all interactions between one gene belonging to one GO term and another gene belonging to the other GO term . Genes belonging to both GO terms were not considered . The coverage is computed as the ratio of the number of tested pairs over the number of possible pairs . For a given number of tested interactions , we computed the expected number of positive and negative interactions , following the global ratio of observed interactions in the full network . We considered the number of observed positive and negative interactions and tested if these numbers significantly differed from those expected using Fisher's Exact Test . We then selected the most monochromatic connections with p<0 . 01 . To group processes into high-level functional categories depicted in Figure 3 , we used the yeast gene annotations provided in [7] where 4 , 414 genes are associated to 18 functional categories . Each biological process is then associated to the functional category that most of its genes are annotated to . To analyze the set of monochromatic processes , we counted the number of processes annotated in each of the high-level functional categories . We then computed the enrichment compared to the background distribution of all processes . The multi-functionality of a gene used in Figure S7 was assessed by the number of processes this gene is involved in , which was computed as the number of GO biological process annotations for each gene restricted to the functionally distinct set of GO terms described in [40] . We used the cellular component part of the Gene Ontology to define protein complexes in yeast . We considered all the children of the GO term macromolecular complex ( GO:0032991 ) . Each term defined a protein complex formed by the genes directly annotated to that term ( not considering IEA annotations ) . This way , we defined 347 complexes encoded by 1 , 795 genes ( Dataset S7 ) . We used this data set in the analysis , unless otherwise stated . We also considered a recent curated consensus of protein complexes in yeast [29] . This Consensus set is a combination of predictions from high-throughput data and curated literature data and consists of 409 complexes . We used two ways to remove the effect of complexes: i ) remove all genes that encode proteins that are part of at least one complex; ii ) remove the interactions that occur between two genes encoding proteins from the same complex , but leaving the genes in place ( in the former case all interactions involving these genes were removed whereas in the latter case only interactions between two genes encoding proteins of the same complex were removed ) . Thus when attempting to explain monochromatic patterns , we considered the following five features , removing either genes or interactions: The overlap between the above features was computed using the Jaccard coefficient . To examine the role of protein complexes at the interaction level , we studied all possible gene pairs . A given pair was considered as involved in a complex if at least one of the genes encodes a protein that is part of a complex . In other words , we partitioned the genes into two classes: 1 ) complex genes ( CG ) : genes that encode a protein that is part of at least one complex and 2 ) non-complex genes ( NCG ) : genes that encode a protein that is not part of any complex . We partitioned the interactions into two classes: 1 ) complex interactions ( CI ) : interactions involving at least one gene of the class CG and 2 ) non-complex interactions ( NCI ) : interactions occurring between two genes of the class NCG . Assuming that the complexes do not significantly affect the structure of the genetic interaction network , we expect the distribution of interaction number among the classes to be the same as the background distribution of all tested pairs . For each interaction class ( CI/NCI ) we computed the ratio of number of observed/expected interactions . The networks were produced using Cytoscape [43] . The position of the monochromatic processes in the GO tree is available as a Cytoscape file in Dataset S8 . | Genetic interactions indicate functional dependencies between genes and are a powerful tool to predict gene function . Functionally related genes tend to have similar profiles of genetic interactions . Recently , global scale mapping of quantitative ( positive and negative ) genetic interactions has been performed . This data clearly shows groups of genes connected by predominantly positive or negative interactions , termed monochromatic groups . These groups often correspond to functional modules , such as biological processes or protein complexes , or connections between modules , but it is not yet known how these patterns globally relate to known functional modules . Here we systematically evaluate the monochromatic nature of known biological processes and their connections in the yeast Saccharomyces cerevisiae . We find that 10% of biological processes and less than 1% of inter-process connections are monochromatic . Further , we show that protein complexes are responsible for a surprisingly large fraction of these monochromatic groups . The monochromatic processes , complexes and connections we find chart a hierarchical and modular map of sensitive and redundant biological systems in the yeast cell that will be useful for gene function prediction and comparison across phenotypes and organisms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetics",
"and",
"genomics/genomics",
"genetics",
"and",
"genomics/functional",
"genomics",
"computational",
"biology/molecular",
"genetics",
"genetics",
"and",
"genomics/bioinformatics",
"computational",
"biology/genomics",
"computational",
"biology",
"genetics",
"and",
"gen... | 2011 | Protein Complexes are Central in the Yeast Genetic Landscape |
Both avian and mammalian basal ganglia are involved in voluntary motor control . In birds , such movements include hopping , perching and flying . Two organizational features that distinguish the songbird basal ganglia are that striatal and pallidal neurons are intermingled , and that neurons dedicated to vocal-motor function are clustered together in a dense cell group known as area X that sits within the surrounding striato-pallidum . This specification allowed us to perform molecular profiling of two striato-pallidal subregions , comparing transcriptional patterns in tissue dedicated to vocal-motor function ( area X ) to those in tissue that contains similar cell types but supports non-vocal behaviors: the striato-pallidum ventral to area X ( VSP ) , our focus here . Since any behavior is likely underpinned by the coordinated actions of many molecules , we constructed gene co-expression networks from microarray data to study large-scale transcriptional patterns in both subregions . Our goal was to investigate any relationship between VSP network structure and singing and identify gene co-expression groups , or modules , found in the VSP but not area X . We observed mild , but surprising , relationships between VSP modules and song spectral features , and found a group of four VSP modules that were highly specific to the region . These modules were unrelated to singing , but were composed of genes involved in many of the same biological processes as those we previously observed in area X-specific singing-related modules . The VSP-specific modules were also enriched for processes disrupted in Parkinson's and Huntington's Diseases . Our results suggest that the activation/inhibition of a single pathway is not sufficient to functionally specify area X versus the VSP and support the notion that molecular processes are not in and of themselves specialized for behavior . Instead , unique interactions between molecular pathways create functional specificity in particular brain regions during distinct behavioral states .
The basal ganglia are a network of subcortical nuclei involved in diverse types of motor function , ranging from simple reflexive habits to deliberated , goal-directed actions . The striatum , the largest basal ganglia nucleus , contains anatomically and functionally separate sub-regions known to mediate distinct forms of motor control and learning in mammals , and investigation of the molecular mechanisms underlying striatal control of learned motor behaviors is an active research area [1] . What has been traditionally called the ‘striatum’ of songbirds such as the zebra finch corresponds to the mammalian striatum , but also contains pallidal neuron types and thus is more appropriately termed the striato-pallidum . As in mammals , the songbird striato-pallidum is active during voluntary movements [2] . In contrast to mammals , the neurons that are dedicated to the learning and production of specific vocal-motor sequences are grouped together in a dense nucleus . This sub-region is called area X and is part of brain-wide circuitry dedicated to song ( Figure 1 ) [2] . This unique specification allowed us to compare transcriptional patterns in tissue dedicated to vocal-motor function ( area X ) to those in tissue that is composed of similar cellular phenotypes but supports non-vocal motor behaviors , such as hopping and wing beating; the striato-pallidum ventral to area X ( VSP ) . Some non-vocal behaviors , such as courtship dances , co-occur with singing [3] , [4] . Comparison of immediate early gene expression between individual cases in which a bird moved but did not sing , versus one that sang but did not make many other movements , indicate activation of the VSP for the former and of area X for the latter [5] . Because of the striking functional contrast between the VSP and area X , we expected to observe different gene expression patterns in these 2 regions during singing . We also hypothesized that VSP gene expression patterns would bear a less significant relationship to singing than those in area X . Since any behavior is likely to be supported by the orchestrated actions of many molecules , as opposed to a single gene or pathway , we used gene microarrays to collect expression data simultaneously from many thousands of genes in both area X and VSP of 27 adult zebra finches . To identify groups of genes with similar mRNA expression profiles across birds , we performed weighted gene co-expression network analysis ( WGCNA ) [6] , a technique that has uncovered patterns of gene co-activity correlated to biological traits and corresponding to functional pathways in multiple species [7] , [8] . In a previous study , we applied WGCNA to microarray data arising only from area X . This resulted in a large network of ∼20 , 000 gene probes from which we were able to identify groups of genes , known as modules , whose expression variability across birds was highly correlated to singing and vocal variability [9] . The co-expression relationships of genes in these area X “song modules” were not preserved in the VSP data , and overall we determined that the more strongly related to singing a given area X module was , the less preserved it was in the VSP . These insights relied on the use of powerful module preservation statistics that evaluated whether modules of a reference network ( the large area X network ) were preserved in the test dataset ( VSP ) . An advantage of these preservation statistics is that they allowed us to rigorously argue that certain modules were not preserved since they did not make use of module assignments in the test data set [10] . But a limitation of our previous study was that we never constructed modules in the VSP . We were thus precluded from identifying any VSP-specific modules , either by computing statistics of VSP module preservation in area X , or by directly comparing the gene composition of modules in each subregion . Our prior results led us to the hypothesis that a VSP co-expression network would share some similarities with area X , but would also exhibit its own unique patterns of transcriptional co-activity . Here we tested this hypothesis by reversing the roles of the area X and VSP data sets: identifying co-expression modules in the VSP and then evaluating their preservation in area X . This approach allowed us to distinguish between different hierarchical transcriptional profiles in adjacent brain regions that control different behaviors but are composed of the same building blocks . Our hypotheses were largely confirmed; a group of 4 VSP modules was well preserved in the new area X network while other modules were highly specific to the VSP , and we did not detect the same strong correlations to the amount of singing as we did in area X . Interestingly , however , we did observe patterns of relatively weak correlations to song features across the VSP modules . These patterns were most distinct with regard to song features that were not highlighted in our previous area X findings , implying that molecular processes outside of proper song control regions may contribute , at least indirectly , to vocal-motor performance . In addition , VSP-specific modules were enriched for processes disrupted in Parkinson's Disease ( PD ) and Huntington's Disease ( HD ) . Because we could directly compare module assignments in the two networks , we were able to determine that genes in two VSP-specific modules were split across multiple area X modules into functionally distinct groups . This strongly supports the notion that single genes and molecular pathways are not in and of themselves specialized for systems-level neural functions or behavior , but instead unique hierarchical patterns of interactions between pathways combine to create functional specificity in particular brain regions under certain conditions .
Animal use was in accordance with NIH guidelines for experiments involving vertebrate animals and approved by the University of California at Los Angeles Chancellor's Institutional Animal Care & Use Committee . Animals and song recording and analysis procedures were identical to those described in [9] . Tissue collection , RNA isolation procedures , and details about the microarrays are described in [9] . Area X and VSP datasets ( raw and processed data ) are available at the Gene Expression Omnibus ( www . ncbi . nlm . nih . gov/geo ) under accession number GSE34819 . Scanned microarray images were analyzed with Feature Extraction Software 9 . 5 . 3 . 1 ( Agilent , protocol GE1-v5_95 and Grid: 019785_D_F_20080327 ) to obtain background subtracted and spatially detrended Processed Signal intensities , which were the input to further data pre-processing . All data pre-processing and co-expression network analysis was done in the freely available statistical software R ( www . r-project . org ) and the WGCNA R library [11] . R functions written by ATH for array pre-processing are available at https://www . ibp . ucla . edu/research/white/code . html . Removal of outlier probes ( including those with sub-background expression levels ) and outlier samples was performed as described in the supplemental information of our previous area X-based study [9] , with the following additions: In the previous analysis , ∼1/2 of the probes on the microarray ( n = 20 , 104 ) were used for WGCNA , and in most cases there were multiple probes for the same gene in the network , e . g . there were 12 probes for FOXP2 . A large portion of the network was also made up of probes for genes whose identity was unknown . In order to streamline the present VSP-based study and ease interpretation of the results , unannotated probes were removed , and one representative probe was selected for each gene using the collapseRows ( ) function , leaving 11 , 482 genes . By default , representative probes are chosen as those with the highest average expression values across samples since they tend to yield the most reproducible results [12] . Co-expression networks were constructed by hierarchically clustering genes based on topological overlap ( TO ) , a biologically meaningful measure of node interconnectedness that compares patterns of gene connection strengths to quantify similarity in the context of the entire network . Other similarity measures such as correlation or Euclidean distance only consider each gene-gene pair in isolation . Thus , TO is effective at filtering spurious or isolated connections , and probes with high TO have increased chances of involvement in the same biological pathways [13] . Modules were defined as branches of the dendrogram obtained from clustering , and labeled by arbitrary colors underneath the dendrogram ( Figure 2A ) . By convention , genes that were not assigned to any module were considered background , i . e . not correlated with genes in the network , and labeled by the color grey . To study module composition , module “eigengenes” ( MEs ) were defined as the 1st principal component of each module , effectively summarizing the expression variability within modules . MEs were used to quantitatively relate gene co-expression patterns to phenotypic traits and construct ME correlation networks to study higher-order relationships among the modules . Module membership ( kME; aka eigengene-based connectivity ) was defined as a gene's correlation to the ME , thus quantifying the extent to which its expression profile conformed to the largest source of variability within the module . Intramodular connectivity ( kIN ) was defined as the sum of a gene's network connections with other module members . WGCNA was performed as described in [9] , using functions in the R WGCNA library [11] , with an additional preliminary filtering step inspired by the procedure used in [14] . Briefly , instead of analyzing a network of all 11 , 482 genes left after pre-processing , an iterative filtering process was performed to enrich the final network with modules composed only of the most densely interconnected genes . The soft threshold for constructing a signed weighted correlation network ( β = 14 ) was determined with the scale-free topology criterion applied to all 11 , 482 genes . A preliminary network was constructed using default module definition ( dynamic tree-cutting ) settings , except for a smaller minimum module size of n = 10 genes [15] . The average TO within each module was defined as the module density , which was then compared to the density of 10 , 000 pseudo-modules of the same size that were generated by randomly selecting genes from the network . A p-value for the density of each module was defined as the number of pseudo-module densities greater than the actual density , divided by 10 , 000 . Genes in modules with p>0 . 01 and grey background genes were removed . The network was rebuilt with the remaining genes , and the process was iterated until all modules passed the density test and there were no more grey genes , leaving 5 , 368 genes in the final VSP network . The area X network was constructed with the same 5 , 368 genes , using the same WGCNA parameter values as in the VSP . No additional filtering was performed in the area X network since it was purposely constructed with reference to the VSP . To confirm the efficacy of the additional filtering for ensuring VSP module robustness , we computed module quality statistics using the WGCNA function modulePreservation ( ) [10] , [11] . Briefly , typical module preservation statistics were used to evaluate the preservation of VSP modules in test networks created by randomly permuting the actual gene module assignments . In this type of comparison , preservation statistics are interpreted as indicators of module density ( how tight the interconnections among the genes in a module are ) and separability ( how distinct modules are from others in the network ) , i . e . module quality . By averaging preservation statistics across many permutations of the original data , module quality statistics are indicative of module robustness and reproducibility . Zsummary scores >10 are interpreted as strong evidence of densely interconnected , distinct , reproducible modules , and scores for the VSP modules ranged from 13 . 2 ( green-yellow ) to 77 . 9 ( turquoise; Figure S1 ) . To enable direct comparisons between the VSP and area X networks and enhance reader-friendliness , we used the WGCNA function matchLabels ( ) to re-assign the area X labels such that modules with significant overlap with a VSP module were assigned the same color label . The function overlapTable ( ) was used to calculate overlap counts and Fisher's exact test p-values for the 2 sets of module assignments , and to produce Figure 3 . The function modulePreservation ( ) was used to compute module preservation statistics , and preservation rankings for VSP modules were determined based on these statistics , as described in [10] . We screened modules for gene markers of neural cell-types and genes associated with PD and HD using lists obtained from [16] and the Ingenuity Knowledge Base ( http://www . ingenuity . com/products/pathways_analysis . html ) , respectively . Enrichment p-values were computed via Fisher's exact test using the 11 , 482 genes remaining after pre-processing as the reference . The same 11 , 482 genes were also used as the reference for enrichment calculations in DAVID 6 . 7 ( david . abcc . ncifcrf . gov ) [17] , which was used for module functional enrichment screening . Lists of genes from each module were converted to Entrez IDs to minimize ambiguity and uploaded to DAVID . Human Entrez IDs were used , as the zebra finch genome remains sparsely annotated , and multiple lines of evidence suggest that the neural systems supporting learned vocalization are highly analogous in humans and zebra finches [18] . For each module , terms with an enrichment p-value<0 . 1 were downloaded for further analysis . Since some modules were enriched for hundreds of terms , many of which could be very similar to one another , we focused our investigations in the following ways: 1 ) To get a broad sense of the enrichment profile for each module , we used DAVID's functional annotation clustering tool to find groups of similar terms , based on genes shared between them , and ranked the clusters by the average enrichment level of terms they contained . 2 ) To prioritize individual enriched terms we first removed any with a false discovery rate ( as computed by DAVID ) greater than 15% , and removed those with fewer than 3 associated genes . Next , we defined a “term significance” ( TS ) score for each term as the average kME of genes annotated by the term , multiplied by 1 – p-value of the term , and ranked terms by these scores [9] . Our intent was to identify biological components and processes that were both highly enriched and represented by genes that most closely conformed to expression perturbations in the module . The functional annotation results are meant to be treated as guides for further investigations and should be interpreted with caution since they represent comparisons of gene lists obtained from avian systems with mammalian databases . Annotation of the zebra finch genome is an ongoing process , making it important to return to our data in the future , including for analyses of probes that are currently unannotated or whose annotations have changed . Results in this paper were based on zebra finch microarray annotations from February 2011 , see http://www . songbirdtranscriptome . net:8080/public . jsp for the most up-to-date annotations . The online Promoter Analysis and Interaction Network Tool ( PAINT; www . dbi . tju . edu/dbi/tools/paint ) [19] was used to identify transcription factor binding sequences ( TFBSs ) overrepresented in subsets of two VSP modules ( blue , yellow ) . Human Entrez IDs for genes in these modules were uploaded to PAINT , where the upstream region ( 2 , 000 base pairs ) of each gene was scanned for TFBSs using the MATCH algorithm and the public TRANSFAC database [20] , [21] . The MATCH filter option “Minimize False Positives” was selected and the parameter “Core similarity threshold” , a measure of the quality of the match between a test sequence and the five most conserved positions in the TFBS position weight matrix , was set to 1 . PAINT can identify TFBSs overrepresented in a subset of genes by comparison to a larger reference set and generate p-values using Fisher's exact test . We examined TFBSs overrepresented in subsets of the VSP blue and yellow modules based on their distribution in the area X network , using the entire module as the reference .
WGCNA of the 5 , 368 genes retained after pre-processing revealed 17 VSP modules ranging in size from 34 to 937 genes ( Figure 2A ) . Computing correlations between the MEs revealed a higher-order network of 3 meta-modules , visible as the 3 largest branches in the dendrogram in Figure 2B–D . Following network construction , we sought to identify any VSP-specific modules . Our previous area X-based study used statistical tests to quantify the likelihood of observing area X modules in the VSP , but we did not build an actual VSP network . Here , in addition to performing statistical tests of VSP module preservation , cross-tabulating module labels between the 2 networks allowed us to explore how functions enriched in VSP modules were split across the new area X network . The re-constructed area X network consisted of 15 modules containing 4 , 022 of the 5 , 368 genes , ranging in size from 14 to 953 genes . The remaining 1 , 346/5 , 368 genes were not correlated strongly enough to genes in any of the 15 modules , and were thus considered background ( denoted by the color grey in the area X module bar under the dendrogram in Figure 2A; also see Table S1 ) . We note that genes referred to as “background” in the context of network construction are those with such weak correlations to other genes that they were not a part of any module . The VSP network was the result of iteratively culling grey background genes until none remained ( see Methods ) , thus the fact that 1 , 346 of these genes were not a part of any module in area X is itself indicative of significant differences between area X and VSP transcription patterns during singing . In the previous study , when genes were selected for WGCNA based on their area X interconnectedness , only 193 of these 1 , 346 genes made it into that much larger original area X network ( 20 , 104 gene probes ) [9] . In other words , these 1 , 346 genes were part of VSP co-expression groups that did not exist in area X , at least after 2 hours of singing . Next , since our primary interest was in studying co-expression patterns found in the VSP but not area X , we investigated which VSP modules these area X grey genes were members of to see if they were co-expressed in functionally significant groups . Module colors in area X were assigned to match the VSP module assignments as closely as possible . This allowed us to compare the overlaps between VSP and area X module membership and identify the VSP module assignments of area X background ( grey ) genes ( Figure 3 ) . We observed significantly large subsets of most VSP modules in at least 1 area X module , e . g . the pink , purple , red , and turquoise modules . This implied a good deal of VSP module preservation in area X , however , some VSP modules were split across several modules in area X , suggesting they were composed of multiple interacting biological pathways that did not similarly interact in area X . As noted above , >1 , 300 genes were not in any proper area X module . The green-yellow , blue , salmon , and yellow VSP modules had the largest proportion of their genes fall into the group of grey background genes in area X ( 48% , 43% , 40% , and 33% , respectively ) . This comprised 60% of the area X grey genes , strongly suggesting that these modules were relatively specific to the VSP . Next , we examined the distribution of VSP module genes across proper area X modules ( i . e . disregarding area X grey genes ) , and for each VSP module , noted its highest proportion of genes in a single area X module . For example , 622 of 937 ( 66% ) genes in the VSP turquoise module were also in the area X turquoise module , and 158 of 817 ( 19% ) genes in the VSP blue module were in the area X brown module ( Figure 3 ) . In support of the idea that the green-yellow , blue , salmon , and yellow modules were relatively specific to the VSP , these 4 modules had lower average proportions of genes in a single proper area X module compared to the rest of the VSP modules ( 25% versus 42% , p = 0 . 017 , Kruskal-Wallis ) . However , the assignment of a set of genes to the same module in both regions does not necessarily mean that the co-expression relationships among those genes are the same , or that any pathway they are a part of is functioning in the same way . Thus , to more closely investigate the preservation of VSP modules in area X , we computed module preservation statistics and derived a composite module preservation ranking for each module . Network statistics that describe co-expression patterns and connection strengths were computed within each VSP module and among the same genes in area X , regardless of their area X module assignment . Module preservation statistics were computed based on comparisons of these network statistics across the 2 regions and VSP modules were ranked based on the outcome [10] . In agreement with the module assignment comparisons , the yellow and blue modules were tied for least preserved , followed by the salmon and green-yellow modules . In contrast , the pink , purple , red , and turquoise modules were 4 of the 7 most preserved , and compared to the rest of the modules , were more preserved on average ( p = 0 . 036 , Kruskal-Wallis ) . Based on the correlation between their MEs , these 4 modules stood apart from the rest of the network in one large meta-module ( Figure 2B ) . This is also apparent in the VSP network dendrogram ( Figure 2A ) . One statistic that figured in the preservation ranking calculations was the correlation between intramodular connectivity ( kIN ) in the VSP and area X . This statistic measures the preservation of intramodular hub gene status across the 2 areas [10] . Since the module preservation statistics were computed with reference to the VSP network , kIN in area X was computed using area X gene expression values , but based on VSP module assignments . Regardless of preservation ranking , normalized gene expression levels in the VSP and area X were nearly perfectly correlated ( Figure 4A–D , I–L ) , whereas kIN values were more weakly correlated in the unpreserved versus preserved modules ( Figure 4E–H , M–P ) . To more directly study the relationship between kIN and differential expression , standard t-tests were used to compare VSP and area X gene expression levels , and the results were compared to VSP and area X kIN ( Figure S2 ) . We found that high kIN is not a good predictor of differential expression , and perhaps counterintuitively , that the region in which a gene had more strong co-expression relationships was also the region in which it had relatively low expression levels . Thus , the differences between the VSP and area X cannot be accounted for by changes in expression levels per se , but are reflected well by co-expression network metrics ( such as the correlation of kIN in the VSP and area X ) that reflect coordinated changes in relative expression levels across groups of interacting genes . To investigate the behavioral significance of the VSP modules , we computed correlations between MEs and 5 continuous singing measurements from each bird: number of motifs sung , pitch , pitch goodness ( a measure of periodicity in the frequency spectrum ) , Wiener entropy ( a measure of the width and uniformity of the spectrum ) , and frequency modulation ( FM ) . We also tested one categorical grouping , whether the bird sang or not , and used the age of the bird at sacrifice as a non-behavioral phenotype for comparison to the singing traits . We did not expect to observe significant correlations between VSP MEs and singing measurements , a prediction that was largely validated; correlations were weak overall ( Figure 5 ) , especially compared to the strength of the correlations we observed in the large area X network in our previous study [9] . After computing the average absolute value of ME correlations to each trait we found that relationships to whether the bird sang or not , the number of motifs sung , and Wiener entropy were qualitatively no greater than to age ( Figure 5 barplot , top ) . This finding supports the functional specification seen in our previous area X study , where these 3 singing traits had highly significant correlations to MEs in 5 singing-related modules in our large area X network , and were predictive of module preservation in the VSP ( stronger correlations = less preserved ) . Thus , we did not expect to find correlations between VSP modules and these song measures . We also re-computed ME correlations to singing and age for the new re-constructed area X modules and found that 3 modules were strongly correlated to the act and/or the amount of singing , replicating part of our previous findings ( brown , midnight-blue , and yellow; these re-constructed area X modules were roughly analogous to the “song modules” in [9]; Figure S3 ) . In addition to these expected findings , we observed relatively strong average ME correlations to FM , pitch , and pitch goodness in the VSP , song features that were not strongly correlated to any MEs in either the previous or re-constructed area X network ( Figure 5 ) . While the VSP ME correlations to FM , pitch , and pitch goodness were still weak compared to previous area X ME correlations to singing , number of motifs , and Wiener entropy , they were organized across the VSP modules in intriguing patterns ( Figure 5 ) . The blue , green , green-yellow , and midnight blue modules as a group showed more negative correlations to FM than the other modules ( p = 0 . 0032 , Kruskal-Wallis ) . Along with the fact that these made up the 2nd of the 3 VSP meta-modules ( Figure 2B ) , this suggests that biological functions they represent may interact in some way related to , or reflective of , the amount of FM in a bird's song . Modules in meta-module 1 ( pink , purple , red , and turquoise ) had more positive correlations to pitch goodness as a group , while meta-modules 2 ( blue , green , green-yellow , and midnight blue ) and 3 ( black , brown , cyan , grey60 , light cyan , magenta , salmon , tan , and yellow ) had more negative correlations ( p = 0 . 013 , Kruskal-Wallis ) , again implying functional groupings possibly related to a song spectral feature . The pitch goodness findings also bore a significant correlation to module preservation rankings; modules that were more preserved ( e . g . those in meta-module 1 ) tended to show increased expression with increasing pitch goodness , i . e . positive correlations , while less preserved modules ( e . g . blue and green-yellow ) showed the opposite pattern ( Figure 6 ) . Finally , correlations to pitch were more positive for modules in meta-module 3 than in meta-modules 1 or 2 ( p = 0 . 0024 , Kruskal-Wallis ) . To identify biological functions and molecular pathways represented in the VSP modules , we used the functional annotation tools available through the Database for Annotation , Visualization , and Integrated Discovery ( DAVID ver . 6 . 7 ) [17] . To focus our analysis of the results from DAVID , we filtered out enriched terms with false discovery rates ( FDR ) >15% and assigned the remaining terms in each module a “term significance” ( TS ) score , which effectively ranked terms by the influence their associated genes had in the module , scaled by the significance level of the term's enrichment ( see Methods; also [9] for a version of TS that accounts for gene correlations to singing ) . We also screened modules for cell type markers and possible disease associations using Ingenuity and gene lists from the literature ( see Methods; http://www . ingenuity . com/products/pathways_analysis . html; [16] ) . Since the main goal of this study was to identify VSP-specific modules that could not have been found in our previous area X analysis , we focused on modules most specific to the VSP: the blue , green-yellow , salmon , and yellow modules . Based on the strength of their ME correlations , the blue and green-yellow modules were more similar to one another than to any other modules in the network ( ME cor = 0 . 67 , Figure 2B ) , and visual inspection of the network dendrogram suggests that the green-yellow module was a relatively distinct subset of the blue module ( Figure 2A ) . A similar relationship existed between the salmon and yellow modules ( ME cor = 0 . 74 , Figure 2B ) , where salmon was a subset of yellow ( Figure 2A ) . Genes in these 4 modules made up the majority of area X background ( grey ) genes ( Figure 3 ) , and based on preservation statistics , they were the least preserved in area X ( Figures 4 and 6 ) . The green-yellow module was highly enriched for gene markers of oligodendrocytes ( genes >10-fold enriched in oligos: p = 1 . 4e−6 , Fisher's exact test ) [16] , e . g . proteolipid protein 1 ( PLP1 ) , fatty acid 2-hydroxylase ( FA2H ) , and myelin basic protein ( MBP ) . Commensurate with this finding , the 5 gene ontology ( GO ) terms in the green-yellow module with the highest TS scores were all related to myelination , including e . g . , GO:0043209∼myelin sheath ( Table S2 ) . 6/8 oligodendrocyte markers from the green-yellow module were grey genes in area X , pointing to distinct co-expression relationships among these genes in the 2 brain regions ( Figure 7A–B ) , suggesting that distinct myelination patterns may contribute to the different behavioral specifications of area X and the VSP in the songbird basal ganglia . Statistical tests of VSP module preservation in human caudate nucleus data [14] revealed that the green-yellow module was moderately preserved in human striatal tissue ( Zsummary = 3 . 34 , 2<Zsummary<10 is considered moderate evidence of preservation [10] ) , suggesting that this module may represent myelination-related pathways important for motor behavior that are conserved within vertebrate basal ganglia . Dopaminergic input from the midbrain modulates the activity of medium spiny neurons in the avian basal ganglia , as in the mammalian basal ganglia [24] , [25] . The death of midbrain dopaminergic cells in humans contributes to altered activity in these circuits resulting in vocal and non-vocal motor symptoms of Parkinson's Disease ( PD ) [26] , [27] . The blue VSP module was highly enriched for genes involved in processes known to be disrupted in PD ( p = 2 . 1e−5 , Fisher's exact test ) , including many implicated in synaptic function and plasticity such as the mu opioid receptor ( OPRM1 ) and growth associated protein 43 ( GAP43 ) . The connections among these blue module PD-associated genes were strikingly different in the area X network ( Figure 7C–D ) , again highlighting distinct gene co-expression relationships in these subregions . GO terms with the highest TS scores in the VSP blue module were related to G-protein signaling , specifically in relation to adenylate cyclase activity and cyclic AMP regulation , phospholipase C activity , and neurogenesis/axonogenesis ( Table S2 ) . Enrichments for most of the same functions were found among the PD-associated genes , with additional enrichments for terms related to behavior , learning , and memory ( Table S2 ) . Genes associated with these terms that had the highest blue module kME scores were OPRM1 , speech-related FOXP2 , 2 glutamate receptors ( GRIA1 and GRIN2B ) , tachykinin precursor 1 ( TAC1 ) , ubiquitin carboxyl-terminal esterase L1 ( UCHL1 ) , catechol-O-methyltransferase ( COMT ) , and phosphatidylethanolamine binding protein 1 ( PEBP1 ) . Both the salmon and yellow VSP modules were enriched for genes associated with processes disrupted in Huntington's Disease ( HD ) which affects basal ganglia function through the death of medium spiny neurons ( p = 0 . 006 and p = 2 . 6e−8 , respectively , Fisher's exact test; Figure 8A–B ) [28] . The same genes showed differential connection patterns in VSP compared to in area X . The yellow module also contained a high number of PD-associated genes ( p = 0 . 049; Figure 8C–D ) . Terms with the highest TS scores in the salmon module were related to protein kinase , alternative splicing , and locomotory behavior ( Table S2 ) . The yellow module contained a significant number of genes that are members of known activity-driven pathways in the Kyoto Encyclopedia of Genes and Genomes database ( KEGG ) [29] , including calcium signaling , MAPK signaling , and long-term potentiation ( Table S2 ) . In agreement with the enriched pathway findings , yellow module terms with the highest TS scores were almost exclusively related to voltage-gated potassium and calcium channel activity , potassium and calcium transport , and learning ( Figure 8E–F ) . When we tested only the yellow module HD- or PD-associated genes , enrichments for the same functions were reiterated , with some interesting additional enriched terms such as GO:0008344∼adult locomotory behavior , GO:0050890∼cognition , GO:0007212∼dopamine receptor signaling pathway , and GO:0008306∼associative learning . The VSP blue and yellow modules were the least preserved in area X . Most genes in the VSP blue module divided into 3 large groups in area X , with the largest number labeled as background ( grey ) , and significant numbers in the brown and black area X modules ( Figure 3 ) . There were clear functional divisions between these groups: genes labeled grey in area X were enriched for catabolic processes and enzyme activation , genes in the area X brown module were enriched for cyclic AMP regulation and neurogenesis , and genes in the area X black module were enriched for steroid biosynthesis , response to steroid stimuli , the cell junction , and membrane-bound vesicles . These findings suggest that functional pathways working together in the VSP blue module , while still important in different area X modules , interact primarily with different sets of genes in area X . These differences may contribute to the contrasting types of motor behaviors supported by area X versus the VSP , for example singing versus nest-building [3] . Much like the blue module , the yellow module was split into 3 main groups in area X , with the largest number of genes labeled as background ( grey ) , and significant numbers in the brown and yellow area X modules ( Figure 3 ) . After examining the enrichments in each of these portions of the VSP yellow module , clear functional distinctions again appeared: genes labeled grey in area X were enriched mostly for potassium channel activity/transport and ATPase activity , genes in the area X brown module were enriched for calcium channel activity/transport and gonadatropin-releasing hormone signaling , and genes in the area X yellow module showed enrichment for genes related to MAPK signaling . Interestingly , the area X yellow and brown modules , which contained significant numbers of VSP yellow and blue module genes , respectively , were highly correlated to singing and specific to area X ( Figure 3 , Figure S3 ) . Since the yellow and blue VSP modules were the most specific to the VSP , this means that the modules that were most specific to each region none-the-less contained some of the same pathways . Presumably , distinct interactions of these pathways with other genes and pathways were what conferred module specificity in each brain region . In the discussion section , we assess the biological functions of key modules and then provide an analogy to illustrate how modular specificity could arise even when these modules share pathways . We also examined transcription factor binding sequences ( TFBSs ) in the upstream sequences of VSP blue and yellow module genes ( see Methods ) . Much like the functional distinctions we observed among subsets of these genes in different area X modules ( and area X grey genes ) , we also found that specific TFBSs were overrepresented in each subset ( Figure S4 ) . The pink , purple , red , and turquoise modules ( meta-module 1; Figure 2 ) were relatively well preserved in area X ( Figures 4 , 6 ) . After performing the same FDR filtering as described for the unpreserved modules , and ranking terms by TS , the most prominent terms in these modules had to do with mitosis/cell division ( purple ) , glycosylation ( purple/turquoise ) , transcriptional regulation ( purple ) , ion transport/binding ( purple/pink/turquoise ) , the extracellular matrix ( purple/turquoise ) , and development ( red/turquoise ) . The turquoise module was enriched for known pathways associated with laminins , collagens and kinases ( KEGG database; See Table S3 for a full list of enriched terms in these modules .
In the songbird telencephalon , neurons that subserve learned vocal-motor communication are clustered together in what are referred to as “song control nuclei” within the larger brain structures of the cortex and basal ganglia . This arrangement permitted us to test the hypothesis that unique patterns of gene activation are associated with a given behavior . Accordingly , in a previous paper , we reported findings from area X where we identified gene modules driven by singing that were not present in the VSP [9] . Because the previous analysis centered on a relatively large area X network ( 20 , 104 gene probes ) , we did not construct an equivalent network in the VSP , and were thus unable to identify any VSP-specific modules , or examine how genes in area X-specific song modules were divided in the VSP . Here , we selected genes for WGCNA based on their interconnectedness in the VSP; comparatively stringent gene filtering criteria resulted in a much smaller network than in the previous analysis ( 5 , 368 versus 20 , 104 gene probes ) , which made it feasible to re-construct an area X network and make direct comparisons of module assignments . This allowed us to compare functional enrichment results in subsets of VSP modules that were split across multiple modules in area X , providing an initial description of how molecular pathways interact differently in the 2 anatomically adjacent , but functionally distinct sub-regions . We did not expect VSP co-expression patterns to relate to singing , an expectation that was met for song measures that were strongly correlated to modules in area X ( e . g . number of motifs sung ) . Surprisingly though , we found that module correlations to pitch goodness were predictive of preservation in area X; less preserved modules tended to contain genes whose expression decreased with increasing pitch goodness , while the opposite was true for more preserved modules . There were also patterns of weak correlations to FM , particularly for modules in meta-module 2 , which included the VSP-specific blue and green-yellow modules . These findings suggest that molecular processes supported by the VSP-specific modules have some bearing on organization of the song frequency spectrum , even though unlike area X , the VSP is not thought to be specialized for vocal-motor learning and production . Alternatively , VSP-singing correlations could reflect changes in body posture or other movements [5] that might affect the quality and modulation of the frequency spectrum , or simply the proximity of the VSP to area X; related molecular activity throughout the surrounding striato-pallidum could be a side-effect of the powerful singing-driven changes seen in area X . The latter seems unlikely however , since the VSP modules were most correlated to the exact song features that did not show significant correlations in area X . We previously predicted that application of WGCNA to expression data from additional song nuclei , e . g . those in cortical areas , would reveal song-regulated gene ensembles not found in neighboring tissue [9] . Relevant to our present findings , studying the relationships between singing and modules specific to tissue neighboring other song regions , e . g . HVC and the outlying nidopallium , could help make sense of the weak module-singing correlations we observed here in the VSP . Cross-subregion comparison of module assignments and statistical tests of VSP module preservation highlighted the blue , green-yellow , salmon , and yellow modules as the most specific to the VSP . The green-yellow module was highly enriched for genes involved in myelination that were not similarly co-expressed in area X , suggesting that distinct myelination patterns may contribute to the functional differences between area X and the VSP , perhaps by shaping the efficiency or temporal precision of axonal communication in neuronal networks that more directly affect behavior . In the VSP , many of the blue module's most significantly enriched functions were focused around G-protein coupled receptor mediated signaling , specifically in relation to intracellular calcium ( adenylate cyclase activity , cyclic AMP regulation , phospholipase C activity ) . In the yellow module , voltage-gated calcium and potassium ion channel activity and MAPK signaling were its most significantly enriched functions . Many genes associated with learning and behavior in these modules code for dopamine , opioid , and glutamate receptors ( AMPA and NMDA-types ) . These enrichments make sense in light of what is known about molecular function in the mammalian striatum , where glutamate receptor activation activates phospholipase C , initiating a host of cellular processes , some of which have a regulatory effect on dopamine signaling . Dopamine in turn modulates glutamatergic inputs to the cell , and coordinated dopamine and glutamate signaling enable corticostriatal long term potentiation ( LTP ) . MAPK signaling is also sensitive to glutamate receptor activity , and MAPK activation may regulate striatal activity and goal-directed behavior through phosphorylation of inwardly rectifying potassium channels [1] . The blue and yellow modules were both enriched for genes associated with processes disrupted in PD and HD , including functions related to synaptic transmission/plasticity , learning , and behavior . Among all of the PD-associated genes in these modules , the gene coding for the mu opioid receptor ( OPRM1 ) was the most highly interconnected , an interesting finding given that opioid receptor signaling is known to affect adenylate cyclase , voltage-gated calcium channels , potassium conductance , transmitter release , MAPK , and protein kinase C [30] . Although cellular pathology and changes in neural activity of the medium spiny neurons have been investigated in genetic models of PD and HD , molecular interactions underlying abnormal motor behavior have not been as well studied , thus our characterization of co-regulation among PD and HD associated genes can provide a baseline for comparison in future work . Furthermore , future studies can explore the significance of the differential patterns of interactions among these genes in area X versus VSP ( Figures 7 , 8 ) , which may convey selective activation of molecular pathways for vocal versus non-vocal motor behavior . Given that these diseases have both vocal and non-vocal motor components , comparison of these modules to those in striatal gene co-expression networks in pathological tissue would help determine which functions and pathway interactions are most disrupted during the disease , including providing insight into interactions between dopaminergic and non-dopaminergic mechanisms . In addition , since there were so many blue and yellow module PD/HD-associated genes contributing to the above biological processes , other genes in the same pathways , or strongly connected to them , could be considered “guilty by association” , and may also play a role in these pathologies [14] . More generally , our findings supported our predictions and speak to the emerging view that massive transcriptional changes can induce shifts between distinct “neurogenomic states” that underlie specific behaviors [31] . Thousands of genes are co-expressed in distinct patterns in area X and the VSP . Specific patterns in area X are highly correlated to the region's specialty , singing , while VSP-specific patterns contain many genes implicated in PD/HD and are involved in processes that are disrupted in these pathologies . For example , gene groups from the VSP yellow and blue modules involved in calcium channel activity ( yellow ) and cyclic AMP regulation ( blue ) were found in the re-constructed area X brown module , which was highly correlated to singing . This means that genes involved in these functions were important in modules that were most specific to their respective striato-pallidal sub-regions ( yellow/blue in VSP , brown in area X ) and had very different relationships to singing . One interpretation of these findings is that functions like calcium channel activity and cyclic AMP regulation are co-regulated with distinct sets of other functions in the VSP versus area X , thus contributing to the behavioral specializations of these areas , an idea supported by the fact that we observed clear functional segregation of VSP yellow and blue module genes across other area X modules . For example , VSP blue module genes related to membrane-bound vesicles were found in the re-constructed area X black module , whereas VSP blue module genes involved in cyclic AMP regulation were found in the area X brown singing-related module . This may imply that within the VSP blue module , the interaction between genes involved in membrane-bound vesicles and those for cyclic AMP regulation is critical for PD-related non-vocal motor function . In contrast , in area X , the cyclic AMP regulation may be less related to membrane bound vesicles but important for vocal-motor function . The following analogy may help make the preceding point more clear . Think of the different biological functions as musicians , the striato-pallidal subregions as performance venues , and the different types of motor behaviors supported by these regions as styles of music . For example , a guitar player , bassist , drummer , and vocalist performing together in one venue play hard rock , but the same drummer and bass player performing alongside a pianist and saxophone player at the other venue might play bebop jazz . The presence of , e . g . , the drummer alone does not determine the style of music being played , nor can a venue specialize in a particular style , e . g . a jazz club , without the right combination of players performing together . Stylistically appropriate types of interaction between particular musicians are what determines the musical form . Thus , for example , the enrichment for cyclic AMP regulation in the VSP blue module does not mean that module is necessarily important for vocal-motor behavior , even though cyclic AMP regulation is also enriched in the area X brown module , which was highly singing-related . It depends on when and where cyclic AMP activity is being regulated , and with what other functions it is regulated in concert with . Together , these results suggest that the expression of a single gene or the activation/inhibition of a single pathway is not sufficient to underlie the functional specificities in area X and the VSP . Better comprehension of the molecular complexity underlying behavior in the basal ganglia will require an understanding of the dynamic interactions between many genes and pathways . WGCNA-type approaches provide snapshots of these interactions at a given time , and can help generate hypotheses about network function and the relative importance of single genes/pathways . | Understanding how gene transcription relates to behavior is challenging . Learned vocal-motor behavior is a complex trait that represents the output of multiple converging genes , pathways , and patterns of neural activity . Here , we applied a systems analytical approach to determine how thousands of genes change their expression levels simultaneously in a region of the vertebrate brain important for vocal-motor function , the basal ganglia , during a specific vocal-motor behavior , singing . We used the zebra finch species of songbird based on similarities between song learning/production and speech , and because they possess a set of brain subregions dedicated to singing . Microarrays were used to measure gene expression levels in one such song-dedicated region and in an adjacent motor area that is not thought to play a role in vocal function . This allowed us to address the question of whether distinct gene co-expression patterns could be found in each area . We found that each area contained unique patterns of transcriptional co-activity , but there were also unexpected overlaps . We conclude that the particular behaviors ( singing versus non-vocal behaviors ) supported by these subregions depend on the particular sets of interactions between molecular pathways that occur in each subregion . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"gene",
"networks",
"neuroscience",
"gene",
"function",
"animal",
"models",
"motor",
"systems",
"model",
"organisms",
"biology",
"neuroethology",
"microarrays",
"systems",
"biology",
"genetics",
"computational",
"biology",
"behavioral",
"neuroscience",
"genetics",
"and",
... | 2012 | Distinct Neurogenomic States in Basal Ganglia Subregions Relate Differently to Singing Behavior in Songbirds |
MicroRNAs ( miRNAs ) are endogenously produced ∼21-nt riboregulators that associate with Argonaute ( Ago ) proteins to direct mRNA cleavage or repress the translation of complementary RNAs . Capturing the molecular mechanisms of miRNA interacting with its target will not only reinforce the understanding of underlying RNA interference but also fuel the design of more effective small-interfering RNA strands . To address this , in the present work the RNA-bound ( Ago-miRNA , Ago-miRNA-target ) and RNA-free Ago forms were analyzed by performing both molecular dynamics simulations and thermodynamic analysis . Based on the principal component analysis results of the simulation trajectories as well as the correlation analysis in fluctuations of residues , we discover that: 1 ) three important ( PAZ , Mid and PIWI ) domains exist in Argonaute which define the global dynamics of the protein; 2 ) the interdomain correlated movements are so crucial for the interaction of Ago-RNAs that they not only facilitate the relaxation of the interactions between residues surrounding the RNA binding channel but also induce certain conformational changes; and 3 ) it is just these conformational changes that expand the cavity of the active site and open putative pathways for both the substrate uptake and product release . In addition , by thermodynamic analysis we also discover that for both the guide RNA 5′-end recognition and the facilitated site-specific cleavage of the target , the presence of two metal ions ( of Mg2+ ) plays a predominant role , and this conclusion is consistent with the observed enzyme catalytic cleavage activity in the ternary complex ( Ago-miRNA-mRNA ) . Our results find that it is the set of arginine amino acids concentrated in the nucleotide-binding channel in Ago , instead of the conventionally-deemed seed base-paring , that makes greater contributions in stabilizing the binding of the nucleic acids to Ago .
As single-stranded RNA molecules of ∼21–23 nucleotide ( nt ) RNAs , microRNAs ( miRNAs ) post-transcriptionally regulate the eukaryotic gene expression by reducing the protein yield from specific target mRNAs , which function is crucial for control of a multitude of critical processes in both plant and animal cells [1] , [2] . Comprising approximately 1% of genes in animals , miRNAs are often highly conserved across a wide range of species [3] . In animals , the functions of miRNA are always associated with regulation of many important processes , including the signaling pathways , apoptosis , metabolism , cardiogenesis and brain development ( reviewed in Ref . 3 ) . In addition , various types of cancers are also discovered as being probably the results of a misregulation of miRNA's expression ( reviewed in Ref . 3 ) . These facts all highlight the importance of miRNAs for both normal cell development and corresponding disease treatment as potential drug target . In both animals and plants , miRNAs recognize their targets through a “seed” sequence . The core of the seed sequence resides between nucleotides 2–7 measured from the 5′-end of the guide strand , which sometimes , to a less extent , still includes the nucleotide 8 [4] . Wealth of data reported for the sequences of animal miRNAs and their targets have illustrated the key role of the seed region of complementarity [1] , [5] . However , some miRNA–target site interactions found in the C . elegans [6] , human cell [7] as well as mice cell [8] often violate the seed rule . It has been revealed that single-nucleotide changes beyond seed and central pairing were important for the miRNA-target recognition in planta [9] . Moreover , the seed match contains both a bulge and a GU wobble , and several other examples of plant miRNA–mRNA interactions with poor seed matches exist [10] . Clearly , only focusing on the seed , even with additional helper parameters , such as the compensatory 3′ pairing and AU-rich sequence occurrence around target sites , is not sufficient , and a better appreciation of the physical chemistry that underlies the target selection is still needed [11] . MiRNA functions by being captured into an argonaute-containing effector complex which is known as RISC ( RNA induced silencing complex ) . As signature component of RISC [12]–[14] , Ago proteins composed of PAZ- and PIWI-containing modules possess a central role in mediating distinct assembly and cleavage steps of the RNA interference catalytic cycle [15] . Structural and biochemical analyses have shown that the ∼130-amino-acid PAZ domain contains an oligonucleotide-binding fold that allows the protein to bind the single-stranded 2-nt 3′ terminal overhangs characteristic of small RNAs processed by Dicer [16] . Actually , the PAZ domain is composed of two subdomains with a cleft in between , where one subdomain consists of a five stranded open β-barrel with two helices on one end of the barrel , and the other one is made up of a β-hairpin followed by an α-helix . These two subdomains have been demonstrated be capable of RNA recognition [17] , yet their dynamic properties involving in the single-stranded nucleic acid binding at atomic level still remain unclear . As to another module of Ago proteins , i . e . , the PIWI domain , it is located at the C-terminus of argonaute across the primary groove from the PAZ domain . Its core fold belongs to the RNase H family of enzymes , containing two highly conserved aspartates on adjacent three β-strands surrounded by α helices [18] . The reported archaeal crystal structure , together with the mutational analysis results , have revealed the necessity of two aspartate residues ( D557 and D669 ) and one histidine residue ( H807 ) for the catalysis of human Ago2 protein [19] . As seen from the crystal structure of Thermus thermophilus argonaute bound to a guide DNA and a target RNA at 3 . 0 Å resolution [20] , a large nucleic-acid-binding channel is observed existing between the PAZ- and PIWI-containing lobes to accommodate the bound ligands . The seed segment adopts an A-helical-like Watson–Crick paired duplex , with both ends of the guide strand anchored in the complex [21] . It is considered that the PAZ and Mid domains of Ago proteins bind to the small RNA 3′ and 5′ ends , respectively , thus in this way they accurately fixed the position of the PIWI-mediated endonucleolytic cleavage of the target mRNA [19] . Recent reports have demonstrated that the RNA-induced silencing complex is an Mg2+-dependent endonuclease in both humans and flies [22] , and the cleavage catalysis is mediated by the PIWI domain of Ago and occurs specifically 10 nt from the 5′-end of the miRNA , leaving the miRNA intact for another round of cleavage [23] , [24] . Very recently , two Mg2+ cations were identified from the crystal structure of Ago complex ( 3F73 . pdb [20] ) , with one cation bound to the catalytic triad ( D478 , D546 and D660 ) of the RNase H fold of the PIWI domain , the catalytic site for mRNA cleavage , yet the other coordinated by Val678 embeded in the Mid domain which might be involved in the anchoring of 5′ end phosphate of miRNA . It is speculated that Mg2+ ion might bind to the RNA substrate through a nonbridging oxygen of the scissile phosphate during catalysis , however , the detailed mechanism concerning the Mg2+ function in the process of RISC-mediated target RNA recognition still remains unknown [23] . In addition , although recent works have provided some insights into the miRNA-target interactions [20] , [21] , several fundamental questions are still open , including particularly: 1 ) how does Ago recognize the miRNA or mRNA dynamically ? 2 ) What causes the conformational changes of miRNA or mRNA within the binding site ? 3 ) How much does each Ago residue or RNA nucleotide contribute to the binding affinity ? 4 ) What are the physical interactions dominating between the target mRNA and the receptor residues ? 5 ) What conformational events and changes in metal binding may occur in proceeding to the transition state from a precatalytic active conformation ? To find possible answers to these questions , in the present work , molecular dynamics ( MD ) simulations were performed on the free Ago ( single ) , miRNA-Ago ( binary ) and miRNA-mRNA-Ago ( ternary ) complexes . The purpose of these large scale simulations is to complement the experiments for better understanding of the miRNA molecular recognition mechanism by providing atomic details that are often inaccessible in experiments due to resolution limits . In addition , to deeply investigate the interactions between miRNA and its target , the principal component analysis ( PCA ) and thermodynamic analyses using molecular mechanics Poisson-Boltzmann surface area ( MM-PBSA ) were also applied . Our modeling and MD results for exploring the dynamic and thermodynamic mechanism of the miRNA-target interaction in the Argonaute protein were reported as follows .
The binding model of miRNA to Ago is extremely similar to that of DNA to Ago in the crystal structures ( 3F73 , 3DLB ) . The nucleotides 1 and 2 at the 5′-end of miRNA are anchored within the binding pocket in the Mid domain , with their phosphate ( base ) oxygens ( nitrogens ) hydrogen-bonded to the side chains of highly conserved residues ( Val678 , Asn445 , His441 and Arg 442 ) as previously observed in the Thermus thermophilus argonaute with 21-base DNA complex ( Supporting Information Figure S1 ( I ) and supporting Dataset S1 ) [20] . The magnesium 681 is coordinated to the first and third phosphates from the 5′ end , as well as the carboxy-terminal carboxylate end ( Val678 ) of the PIWI domain ( Figure S1 ( II ) and supporting Dataset S1 ) , and the second magnesium 679 is also well located in the catalytic triad ( Figure S1 ( III ) , supporting Dataset S1 ) [20] , [21] . Nucleotides 22 and 23 at the 3′ end of the miRNA are anchored within the binding pocket in the PAZ domain ( Figure S1 ( IV ) and ( V ) ) , with the oxygens of the phosphate linking base 22 and Arg228 , and three H-bonds formed with Trp239 , Glu206 , Pro208 ( Figure S1 ( IV ) and supporting Dataset S1 ) . And the 3-OH of nucleotide 23 hydrogen-bonded to acidic Ser218 side chain , as previously observed in PAZ–siRNA [20] , [21] and PAZ–single stranded RNA complexes ( Figure S1 ( IV ) and supporting Dataset S1 ) [20] , [21] . In addition , there are extensive hydrogen bonding and salt bridge formations between the backbone phosphates of the miRNA strand and the nucleic-acid-binding channel in the protein ( Dataset S1 ) and proposed models of the catalytic cycle [20] , [21] . In the ternary system , the bound mRNA strand threads its way within a central channel in the Mid and PIWI-containing ( Mid and PIWI ) lobes of the bilobal scaffold of Ago , with its segments 2–8 ( seed ) hydrogen-bonded to miRNA ( Figures S1 ( I ) and ( V ) ) . The bases are splayed apart at the nucleotides 1–2 from 3′ end , with base 1 stretched over the plane of the binding site and the last 8 nts stretched out from the PIWI groove which is almost perpendicular to the miRNA line . These results suggest that our binary and ternary complex models derived from docking are reliable . For all three systems , the temperature , total energy , mass density , and volume are found relatively stable throughout the MD simulations . The conformational drift of Ago structure is also measured in terms of the root-mean-square deviation ( RMSD ) with respect to the starting structure ( supporting Figure S2 ) . Plotting the RMSD of all Cα atoms as a function of time for the three simulation copies reveals relatively small changes in the structure , indicative of satisfactory simulations performed ( Figure S2 ) . The RMSD values ( Figure S2 ) observed for Ago show a relatively wider range ( 4 . 15±0 . 63 Å ) in the stable time , suggesting that significant domain movements are involved . Whereas the RMSDs of Ago bound with guide RNA vary within a much narrower window of 2 . 72±0 . 31 Å and , in all cases , appear more stable , apparently due to the presence of the binding to miRNA . Notably , for the ternary complex , the RMSDs observed restore to a wider range ( 2 . 80±0 . 52 Å ) similar to that of the free Ago , suggesting that this complex possesses more fluctuations after binding to the second RNA ( mRNA ) . Actually , this increased structural flexibility of the ternary system is attributed to the existence of more flexible subdomains such as the loop 1 ( amino acids: 198–239 ) and sheet 1 ( amino acids: 74–102 ) ( see following section E of Results ) , in accordance with earlier observations [21] , [22] . The visual analysis of each trajectory obtained from these three simulations also supports above observations . These phenomena imply that for Ago , its binding to the guide RNA greatly increases the complex's stability , yet when the binary complex further binds to another target RNA , the stability of the complex is slightly decreased due to the perturbation of the free 5′ end of mRNA ( see more discussion below ) . The B-value [25] calculated from all-atom MD simulation provides another approach to evaluate the convergence of the dynamical properties of the system . In the present work , the obtained normalized B-value [25] results ( Figure 1 ) well agree with those reported in the X-ray structure ( 3F73 . pdb [20] ) . In particular , the simulations reproduced the sharp peaks observed in the crystallographic structure around loop 1 and sheet 1 ( supporting Figure S3 ) . In addition , all those residues with higher fluctuation values are found belonging to the highly mobile solvent-exposed amino acids ( loop 1 and sheet 1 ) in the PAZ domain . On the contrary , only a small degree of flexibility is observed at the Mid domain ( 326–462 ) . As to the PIWI domain ( amino acids: 463–678 ) , it also exhibits much larger stability than the PAZ domain , which is consistent with its structural role in the heterotrimer [20] . To extend this analysis , a principal components analysis [26] was performed on the equilibrated portion of relevant trajectories , ending in results agreeing well with the B-value analysis in that the principal component contains the motion of loop 1 and sheet 1 region . And this region appears to be significant and be modulated by the presence of bound miRNA within the central cavity . This consistency between the simulation results and experimental observations proves the reasonability and validity of the models . Principal component analysis provides one way of extracting large-scale motions in proteins . Similar in spirit to a normal-mode analysis , PCA breaks up the total motion into contributions , each with a pattern of coherent motion [26] . In this study , the simulation trajectories of each system were analyzed for dominant collective displacements using PCA of the fluctuations of Cα atoms [27] . The first two eigenvectors in the ternary model from the PCA capture about 55% of the variance of entire crystallographic ensemble and thus represent large-scale collective motions , with subsequent eigenvectors capturing significantly less fluctuations ( see Supporting tables ( Table S1 ) ) . It is thought that the large displacements seen in the first few eigenvectors of such analyses represent functionally important global movements , while lower-order eigenvectors are smaller , localized fluctuations that do not influence the function [27] . Accordingly , PCA studies focus on these dominant motions and Figure 2 shows the projection of each member of the crystallographic ensemble onto the plane defined by the top two eigenvectors . On these projections , clusters of stable states were observed with two apparent features from these plots . Firstly , the clusters are well defined in all three systems , indicating that these systems sample two ( for the single and ternary systems ) or three ( for the binary system ) distinct minima during the molecular dynamics trajectory . Secondly , the single system is found covering the largest region of the phase space , the ternary system covers the second largest region and the binary system covers the smallest area . By analyzing the structural clusters , the significant shift seen between free Ago and Ago ternary complex clusters is found mainly resulted from large conformational change in PAZ domain ( see more discussion below ) . Our observation thus corroborates with the idea that free Ago has higher flexibility than the bound Ago with RNAs at room temperature . To characterize the collective motions represented by the dominant eigenvectors , interpolations between the extreme projections of the crystal structures are provided in section E ( in the results ) , with purpose to convenient compare with following results of section D . In order to reveal a clear-cut structural difference of the RNAs in Ago binary and ternary complexes , PCA was also performed on each trajectory of the two complexes independently . The top three eigenvalues for miRNA in the binary system explain 56 . 8% of the total variance , and 70 . 1% of that in the ternary system ( Supporting tables ( Table S2 ) ) . As for mRNA , in the ternary system 81 . 3% of variance is represented by the first three dominant eigenvectors ( Supporting tables ( Table S3 ) ) . Figure 3 shows the superposition of extreme projections on PC1 of structures of the guide RNA and its target . The bulk of the displacement is seen in the 5′-end , involving five nucleotides in mRNA , and eigenvector 1 represents a concerted swing event in the solvent environment . For miRNA , no evident conformational variations are found , which is shown by the good superposition of the extreme structures in the binary system , and the same story holds for the ternary system ( Figure 3 ) . However , the comparison of the conformations of miRNA between the binary and ternary complexes shows a significant change in segments 15–23 from 5′-end . This part of miRNA rotates in a clockwise direction up to 50° and translates ∼5 Å from the binary to ternary complexes in the presence of its target , representing the most open state of the cleft ( PAZ binding pocket ) . Widening of the PAZ-PIWI distance in the substrate-bound forms represents the accommodation of nucleic acid into the RNA binding channel . Eigenvector 2 represents a significantly smaller displacement of atoms and shows quite similar results as PC1 and is thus not discussed here for space saving . Clearly , the present miRNA-target exhibits good pairing to the 5′ end of the miRNA as shown in Figure 3 ( upper part ) , no evident interaction is found for base-pairing to the 5′-end of the mRNA with its guide RNA ( see supporting Figure S4 and Dataset S2 ) , failing to meet the demand of a canonical site rule , i . e . , it has enhanced 3′ pairing in addition to a sufficient 5′ seed [28] . Therefore , the functional importance of the 5′ end of mRNA might be related to other physical characteristics or functions such as the dynamic properties in the binding complex . We could speculate that , due to the violent sway of 5′-end hinged by nucleotide 18 , the cleaved part of mRNA can be easily disassociated from the Ago system , thus accelerating the catalysis cycle of miRNA . As the RNA binding channel of Ago consists of several conserved motifs , the Ago conformational switch probably induces the movements of the bound RNAs . To reveal the correlations , the motion modes of Ago in each structure projected in PC1 and PC2 identified by DYNDOM [29] are presented and shown in Figure 4 . PCA results show that the dynamics of the single , binary and ternary systems appear to occur mainly along the first and second principal components , the subspace of which accounts for , 57 . 9% , 34 . 6% and 54 . 4% of the overall motion , respectively ( Supporting tables ( Table S1 ) ) . As for free Ago , DYNDOM analysis discovered two key amino acid segments related to the motion of the first principle component , viz . , loop 1 and sheet 1 , two highly conserved subdomains in PAZ domain ( Figure 4A ) . Loop 1 shows an independent movement of helical twist by hinge 1 , and sheet 1 exhibits as a dominant contributor among the residues involved in the interdomain rotation and bending . And the large scale motions of loop 1 are also captured by PC2 with respect to domain R2 ( Figure 4B ) . These results indicate the existence of a reasonable agreement between the X-ray sub-domains ( 3F73 , 3DLB ) and those in the PCA eigenvectors , particularly those in eigenvectors 1 and 2 . Therefore , we speculate that the large scale motions of PAZ play a mechanical role for the Ago binding activity , resulting in the generation of a wider and shorter nucleic acid binding channel necessary to hold and orient the bound RNA guide strand . In Ago binary complex , the eigenvector explaining the largest part for the variance of the trajectory of Ago corresponds to the deformation in which a conformation containing two large subdomains ( the blue , B3 , and red , R3 in Figure 4C ) forming a crescent-shaped base . In the binary complex , the first eigenvector represents a deformation which opens and closes the boat-like shape of the nucleic acid binding channel ( Figure 4C ) , connected by a stalk-like linker region between the N-terminal and the PAZ domains , an interdomain connector cradles the structure . The second eigenvector displays a pivotal rotation of the subdomain loop 1 , and the conformational changes are extended to the base PIWI-containing lobes ( Figure 4D , with detailed residue numbers seen in Supporting tables ( Table S4 ) ) . Clearly , this subdomain movement of the protein effectively adjusts the RNA positioning and , in this way , avoids the clashes with other parts of the full-length protein , which also provides a correct position for the binding to target RNA . In mechanistic terms , we favor the view that the conformational changes in domains B3 and R3 ( Figure 4C ) and associated sliding and flipping of the miRNA-strand ( Figure 3 ) are triggered by widening of the substrate-binding channel between the PIWI and N domains to accommodate a lengthening of the target RNA complex . Such changes not only push the PAZ domain away but also release the 3′ end of guide strand from the PAZ-binding pocket ( Figure 4C ) . During our simulation , in most of the cases , the simple stretching and twist movements for such large subdomains ( half of the protein ) of Ago display fewer fluctuations of atomic coordinates . This is well consistent with the RMSD plot of the binary complex , which is the most stable one among the three systems . It is likely that the first and second motion modes for PAZ domain in miRNA-Ago complex are related to the miRNA 3′ end conformational switch , because the fluctuations of the eigenvalues corresponding to these two components are in good accordance with the conformational change profile of miRNA ( Figure 3 and Supporting tables ( Table S1 ) ) . In the ternary system ( Figure 4 , E and F ) , the first two animated eigenvectors of Ago correspond to the stretching motion of a helical fragment around loop 1 in PAZ domain , which is similar to the previous one in free Ago , except for the immobilization of sheet 1 relative to PIWI domain . Half of the residues ( loop 1 ) in this case have more pronounced contributions to the fluctuations of the atomic coordinates . In addition , Ago also shows a twist movement around the linker region of the protein ( Hinges 10 and 11 ) , which might explain why miRNA shows different conformations in the binary and ternary systems ( Figure 3 ) . Thus , we conclude that the first and second components may truly correlate with the conformational dynamics of miRNA in the binding channel of Ago . No motifs are found to be likely related to the target RNA conformational change , which are in good accordance with the conformational change profile of mRNA ( Figure 3 ) . The comparison of the PCA results of free Ago and Ago ternary complex further reveals that the first component of free loop 1 is highly similar to that of loop 1-miRNA ( Figures 4 , A and D ) , suggesting that it is a fundamental motion of loop 1 . The extreme conformations of PC1-PC2 are further used as input to detect the motion hinges [30] as well as to quantitatively analyze the magnitude of the protein movement . Table 1 shows the hinge parameters for the first two PCs , and each hinge is indicated in Figure 4 , A–F . From this table , we can find that hinges 1 and 8 ( Figure 4 , A and E ) have the largest magnitude of rotation of about 25° in all hinges , making the PAZ domain partially open in the single and ternary complexes; And the biggest closure movement is subjected to loop 1 and sheet 1 , indicated by hinges 6 and 10 of ∼100% closure degree in axis direction . The complicated and large movements of PAZ domain around these hinges imply its functional importance . In the binary system , the relative opening-closure degree of two subdomains ( B3 and R3 ) is represented by hinges 5 , 6 and 7 , showing a trend of lengthening and widening of the nucleotide-binding channel ( Figure 4 ) . In addition , another interesting finding is that five out of eleven hinges ( 1 , 3 , 6 , 8 and 11 ) are relevant to loop 1 , the main part of Ago , which might further demonstrates the predominant role of loop 1 in Ago for the RNA binding . In addition , to compare the conformational changes in different systems , the average geometries of Ago from each of the simulations are structurally superimposed ( Figure 5 ) . It can be seen that in free Ago ( blue ) , the PAZ domain approaches most to the geometric center of left subdomain ( amino acids: 310–678 ) ( Figure 5 left ) , with a distance of 33 . 4 Å between the two geometric centers of the Cα atoms of left and right subdomains ( amino acids: 1–310 ) and an entrance area of 396 Å2 ( right cross-section ) of PAZ domain ( Figure 5 right , in blue color ) . After binding to a miRNA , this nucleic-acid-binding channel between the PAZ- and PIWI-containing lobes of argonaute is slightly lengthened to 34 . 9 Å , but the entrance significantly shrinks to ∼299 . 4 Å2 ( Figure 5 right , red color ) . And the channel can be further greatly lengthened up to 41 . 0 Å after binding to another target RNA strand , with the PAZ entrance being ∼25 . 4 Å2 more narrowed compared to the binary system ( Figure 5 right , green color ) . From the above conformation studies on the three Ago structures , an unexpected mechanistic insight emerges that the short and wide binding channel in the free state is able to relieve the topological stress of Ago on the guide RNA , and make the RNA 5′-head easily attach to the 5′-phosphate-recognizing Mid binding pocket from the entrance of PAZ domain [20] , [31] . In order to understand the importance of base paring on the stability and flexibility of the seed ( bases 2–8 ) , the all-atom RMSDs of the seed bases of miRNA in free and bound states are evaluated ( supporting Figure S5 ) . It is observed that the segment in unbound state undergoes larger fluctuations , increased up to 6 times compared to that in the state bound to a target RNA ( RMSD = ∼1 . 5 Å , see supporting Figure S5 ) . The 2–8 segments of the bound guide miRNA form a stacked array , such that the solvent-exposed Watson–Crick edges of bases 2 to 8 in the ‘seed’ segment are positioned for nucleation with target RNA [20] . Therefore , we conclude that large fluctuations of unbound bases should be of functional significance , which could increase the possibility of formation of the seed duplex by base-paring H-bonding interactions . In addition , the conformational changes of two bases 9 and 10 proximal to the seed segments , which might be crucial for the catalytic cleavage of target RNA by PIWI domain , are also analyzed [20] . Figure 6 shows the RMSDs of distances of four inter-base-pair-hydrogen bonds and dihedral angles ( base–base ring ) for A7-U7′ , C8-G8′ , U9-A9′ and A10-U′10 versus simulation time . As expected , the distances of A7-U7′ and C8-G8′ ( seed ) are less than 3 . 0 Å over the entire trajectory , indicating that the structure of the two-base-pair dimmer is stable . Interestingly , for U9-A9′ , in the fist 0 . 2 ns , H-bonds are formed and subsequently break and the distance between two corresponding bases increases up to 7 . 5 Å until to 20 ns , afterward , this distance descends to ∼3 . 0 Å and the H-bonds form again , but at 15 ns , the H-bonds break again and remain broken until to the end . This result reveals that nucleotide 9 may switch between two kinds of conformations in the binding site ( Figure 3 ) , which might explain why there is a 7- or 8-mer seed match in different species , except for the conserved 6 mers [4] , [32] . During the whole simulation process , nucleotides 10-10′ are found undergoing larger conformational changes . The H-bond distance remains below ∼3 . 0 Å during the first 1 . 0 ns , and reaches ∼6 . 5 Å in the following 3 . 0 ns . Afterward , it descends but still remains large ( 5 . 0–6 . 0 Å ) . The dihedral angle profile describing the base-base rings of the four base-pairs exactly matches the H-bonding profiles ( Figure 6 ) , in that the base-pair dihedral angles for rotation around the A7:C2-N1-U7′:N3-C2 ( abbreviated as T1 ) and C8:C2-N3-G8′:C2-N1 ( T2 ) bonds are planar and oscillate around 20° or −20° , being close to the optimal geometry of isolated seed segment . These data indicate that the hydrogen bonds formed between 7-7′ , 8-8′ base pairs are stable ( 3 . 0 Å ) . The U9:C2-N3-A9′:N6-C6 ( T3 ) bonds show the same pattern as T1 and T2 from the beginning to 15 ns , but deviates for the last 5 ns , further supporting the switch of conformations of U9 in the whole trajectory . Different from above T1 and T2 , the dihedral angle for 10-10′ ( T4 ) remains about 170° in the first 2 ns , and then is inverted to 120° for most of the last trajectory , leading to the disruption of the initial H-bonding . This computed binding mode is identical to the local geometry reported in the crystal structure ( 3F73 ) , i . e . , the base 10-10′ equilibrates to a departure conformation , which makes the phosphate diester bond of U10′ approach the catalytic site in PIWI domain , so that the cleavage event can take place ( see more discussion in section G ) . These findings might also explain , from an atomic level , why the repression of gene expression was not affected by a mismatch at position 10 for miRNA and target [28] , since the two bases 10-10′ do not have to be complementary and cemented together for target recognition . The X-ray crystal structure of Ago complex indicated that an array of water molecules play an important role for the receptor-mRNA binding ( 3F73 , 3DLB , 3HM9 ) . To study the specific roles of water molecules in the binding of mRNA to Ago complex and in the conformational change of seed site , we further performed a detailed analysis on the MD trajectory of the ternary complex using the VMD program [33] . Figure 7 shows one representative snapshot isolated from the MD trajectory . By dynamic simulation of the ternary system , it is revealed that the hydration around A7-U7′ and C8-G8′ is not defined , while the baseparing of U9-A9′ and A10-U10′ is complemented by a highly specific hydration site occupied by one water molecule . Because of poor accessibility , this site is vacant during the initial 7 ns of simulation , but at 7 . 0 ns into the simulation it becomes hydrated and remains occupied for the rest of the simulation . And this hydration site forms five H-bonds with U9 , A9′ and A10 as shown in Figure 7 . This observation confirms the above H-bonding picture for A10-U10′ , where after 7 . 0 ns the distance of the two bases becomes stablized and enlarged to 6 . 0 Å ( Figure 6 ) , thus a relatively deep pocket between the two bases is formed to accommodate a water molecule . This hydration site shows residence time of the water molecules in a long range of 15 ns . Therefore we speculate that the long-residing and presumably tightly bound water molecule contributes considerably to the stability of U10′ in the catalytic binding site in PIWI domain . In the crystal structure of Ago complex ( 3F73 ) , a well-ordered divalent metal ion ( Mg2+ ) is bound to the C-terminal carboxyl groups of triad D474 , D541 and D653 at the PIWI domain [20] . In order to investigate the physical function of this ion in the PIWI active site of the ternary complex , possible hydration effects and ion interactions were investigated . An analysis of the water density around the PIWI catalytic site reveals several highly occupied hydration sites ( Figure 8 ) . Six hydration sites are observed in the binding pocket near the Mg2+ in both binary and ternary complexes referenced as H1–H6 in the following text . In the binary complex , three of these sites ( H2 , H3 and H4 ) are above a plane formed by the triad carboxyl groups , with first site ( H1 ) below this plane interacting with the two O ( F473 ) and O ( L649 ) atoms ( Figure 8 ) . With the abolishment of hydration sites H1 , H3 ( D474 ) and H4 ( D653 ) in the ternary structure , two new hydrations sites H5 and H6 are formed by OD2 ( D474 ) and OD1 ( D541 ) , working as glue to stabilize the catalytic center of the tetrahedron structure . These hydration sites appear to compensate for the crystal waters in 3F73 [20] , in which only sixteen irrelevant water molecules were crystallized . Hydration sites show residence time of individual water molecules in the long range of 0 . 1–20 ns except H3 and H4 which occupancy is only around 50% of the entire trajectory ( see Supporting tables ( Table S5 ) ) . The long-residing and tightly bound water molecule ( Table 2 ) in H2 is functionally significant , facilitating the RNA hydrolysis during catalytic cleavage by RNase-H-containing nucleases together with the Mg2+ cation [34] , although experimentally , at present it is not possible to identify the position of the water molecule that would participate in and be positioned for in-line attack on the scissile phosphate target . Interestingly , the transition from binary to ternary binding mode of the system does not cause any water switch in H2 but results in the vanish of H1 . In the binary structure , as schematized by the sketch map on Figure 8A , the four oxygen atoms OD2 ( D474 ) , OD1 ( D541 ) and OD1 , OD2 ( D653 ) form a ‘strict’ plane , indicating by the sum of angles between the triad residues of 359 . 3° . Mg2+ is well located at the center of the plane of the four-membered ring . However , it is likely that the planar structure is rather unstable due to the short stacking distance and effect of oxygen–oxygen electronic repulsion . Previous work has found that the metal ion bound to the C terminus observed in both the RNA-bound and free forms of AfPiwi was not observed in PfAgo , even when crystals were soaked with Mn2+ , the possible reason is that the single metal ion can not keep stabilized in the plane without additional help from outside of the triad [19] . Based on this thought as well as the simulation results , we then found out the particular significance of the well circumscribed hydration site H1 . The water molecule occupying this site stabilizes the planar state of the Mg2+ ion mediated-tetramer by a strong water-ion electrostatic attraction , with its two hydrogen atoms hydrogen-bonded to the two adjacent carbonyl groups of F473 ( H-bond distance is 2 . 29 Å ) and L649 ( 2 . 32 Å ) utilizing their acceptor abilities , respectively . The stability is confirmed by the RMSD plot of the Mg2+ with interfaces 7 atoms ( Mg2+ , four O atoms , and two water molecules at H1 and H2 sites ) in the binary structure ( Figure 8A ) , with an average RMSD less than 1 . 5 Å ( see supporting Figure S6 ) . In addition , the results of distances between these atoms are in the range of 1 . 8 to 3 . 1 Å ( Table 2 ) , also supporting the conclusion that the H-bonds in this system are stable . However , this geometry is destroyed after the binding of a target RNA . Due to the attraction from O1 ( P ) -RU717 on Mg2+ ( Figure 8B ) , the basal plane is drawn away from the axial direction for 3 . 0 Å , and the metal ion is further pulled out of the DDD-constructed plane for 1 . 2 Å . This results in a regular tetrahedron structure ( symmetrical ) with the ion as the vertex , as revealed by the three similar dihedral angles in size ( ∼90° ) ( Figure 8B ) . The RMSD plot for this structure in entire trajectory ( <3 Å , see supporting Figure S6 ) and the key distances between the involved atoms ( Table 2 ) also demonstrate that this catalytic center in PIWI domain is quite rigid . Clearly , the movement of Mg2+ occurring upon binding of the target strand could avoid the clashes with other parts of the protein and bring the target RNA in correct position for cleavage . This finding also explains why the H1 is absent in the ternary structure , for: 1 ) no stable H-bond can be formed as in the binary complex because the Mg2+ is drawn too far from the two residues F473 and L649; 2 ) The pocket under the triad plane built by F473 and L649 becomes so much narrower in the ternary complex that no water molecule can easily enter . In addition to the H-bond network for miRNA-mRNA ( seed ) discussed above , we also examined the hydrogen bonds between the mRNA and Ago protein . Examined 3000 snapshot structures for the last 12 . 0 ns , we identified direct hydrogen bonds . Base 1′ is splayed and hydrogen-bonded to O of F640 with the backbone phosphate group hydrogen-bonded to the side chain ( 2HH1 , 2HH2 ) of K440 . Base 1 is the only residue before step 9 ( here , step 1 is referred to the 1st base counted from the miRNA 5′ end , step 2 is the 2nd base of miRNA 5′ end , and so on for others . ) on the target strand that makes base-specific contact with the Ago scaffold , and this observation is consistent with the reported sorting of small RNAs in Ago complexes [20] . A helix from the Mid domain appears to separate the miRNA and its target strand at the 5′ end , which state of separation is stable in the entire simulation time shown by the dynamics simulations . This unpairing of the first putative guide-target ‘basepair’ is in agreement with the observation that base pairing at this position does not appear to be required for target recognition [35] . The guide DNA–target RNA seed duplex spans positions 2 to 8 , with the scissile phosphate ( 10–11 steps ) on the target strand positioned opposite to the catalytic residues ( D474 , D541 and D653 ) of the RNase Hfold of the PIWI domain [20] . For steps 11–23 , the basepairing is disrupted , and the target RNA subsequently runs in an almost perpendicular trajectory to the guide RNA , stretching out from the crescent base ( Figure 3 and supporting Figure S4 ) . As seen from Figure 9 , Ago binds directly to the middle-last part in the GAAUCAAU region of the sequence ( 5′-ACAGCAGAAUCAAUAGUCUUCCG ) . From segments 10–17 , the plane of mRNA remains parallel to the wall of the groove and forms eleven H-bonds with seven amino acids in the ternary Ago complex ( ( U10′-Lys657 ( 2 H-bonds ) and Lys570 ( 2 ) ; A11′-Lys570 ( 2 ) and Arg543 ( 2 ) ; A12′-Arg569 ( 1 ) ; C13′-Lys611 ( 1 ) ; U14′-Arg102 ( 1 ) ) . This interaction is stabilized by the fit of the skeleton phosphate oxygen hydrogen-bonded to the guanidinium group of arginine and the ε-amino group of lysine along the narrow groove of PIWI . The six 5′ -end overhanging nucleotides ( 18–23 ) that are not adjacent to the binding channel freely stretch in the solution . Interestingly , we found that these key residues interacting with mRNA are all basic , building a positively charged channel for mRNA binding , as shown by the blue surface in Figure 9 ( see detailed discussion in section Thermodynamic Analysis ) . This might compensate for the unavailability of the Ago/miRNA/mRNA complex , providing new information for exploring the interactions of the ternary complex in the mRNA binding channel since the bases of the target RNA can not be fully traced in the present X-ray structures ( 3F73 . pdb , 3DLB . pdb ) . As a complement to the structural data discussed above , presently the estimates of the binding free energy of the guide and target RNAs to Ago in various binding modes were performed . For space reasons , only the main points of our analysis were outlined below . Table 3 shows the results of a typical set of MM/PBSA calculations for the binding of miRNA to Ago and mRNA to miRNA-Ago complex , where the five energy terms in Eqs . ( 3 ) and ( 4 ) in Materials and Methods , as well as the total binding energy and the experimental results are provided . It can be seen that the electrostatics , van der Waals , and nonpolar solvation terms are favorable , whereas the polar solvation and entropy terms are unfavorable for the binding in both two complexes . Since the RNAs have a net negative charge , the energies are dominated by the electrostatics and polar solvation terms , which two energies can not even cancel , as can be seen in the fifth row from the bottom of Table 3 . This is a manifestation of the dielectric screening of the solvent . Actually recently it has been argued that the net binding of RNA to protein is dominated by this term [36] . In addition , the van der Waals term is also rather large ( −187 to −330 kcal/mol ) , correlating with the binding areas to protein of the molecule . But the nonpolar solvation term is small for the two molecules ( −29 to −49 kcal/mol ) . As to the entropy term , it is intermediate in size ( 82–123 kcal/mol ) , for the entropy is dominated by the nearly constant translational ( ∼12% ) and rotational contributions ( ∼13% ) , whereas the variation is caused by the vibrational contribution ( ∼75% ) . Concerning with the calculated changes in solute entropy , i . e . , the TΔS , they are physically reasonable , since the flexible RNA molecules often exhibit as large as even up to 360 kcal/mol entropy due to large vibrations [37] . In general , the total estimated binding free energy ΔGbind fall in the profile of the observational results [38] , [39] . However , from these results we have to confess that the standard deviations of each component of the MM-PBSA are relatively huge , which is caused by the big vibration other than the translation and rotation movements for such a big molecular system . Thus this propagation of errors might finally influence the accuracy of the total free energy obtained . In order to fully investigate the influence of the residues in protein on the interaction of binders , the RNA-residue interactions in each binary and ternary complexes were decomposed and compared systematically . Figure 10 shows the contribution of each residue and base to total ΔGbind-miR . For the miRNA binding to free Ago , the favorable residues can be classified into four clusters around residues R47 , R228 , R442 and R644 , belonging to the PAZ , Mid and PIWI domains of Ago ( Figure 10 ) . Only those residues whose respective contribution to ΔGbind-miR is above 1 . 4% are listed here , a threshold that ensures 80% of total ΔGbind-miR is contributed by the selected residues . Very surprisingly , we find that the involved 16 amino acid residues ( Arg47 , Arg55 , Arg77 , Arg168 , Arg188 , Arg190 , Arg228 , Arg232 , Arg282 , Arg442 , Arg543 , Arg575 , Arg604 , Arg608 , Arg644 and Arg654 ) are all arginine , which account for 20% in all 80 arginines in Ago protein ( Figure 10 ) . We also observe that there are extensive hydrogen bonding and salt bridge formations between the backbone phosphates of the guide RNA strand . As expected , these Arg side chains uniformly span the binding channel in PAZ and Mid domains but are more concentrated in the PIWI domain ( 6 arginine molecules ) ( Figure 10 ) . This makes the catalytic site in PIWI domain more basic , thus facilitating the binding of a RNA guide sequence , which supports the previous assumption about the function of the RNA-binding groove [17] . After the discussion of contributions from amino acids , another important issue to address is to assess at what level basepairings contribute meaningfully to their function by thermodynamic analysis . Two most important bases are defined , i . e . , 1st and 3rd nts from 5′ end , accounting for ∼15% of the total energy ( ΔGbind-miR ) ( Figure 10 ) . The first base of the guide RNA is flipped out and stacked with an Arg side chain , and the 5′-phosphate is placed in a highly conserved pocket in the Mid domain interacting with Val678 , Asn445 , His441 and Arg442 as previously observed in the Thermus thermophilus argonaute with 21-base DNA complex [20] . Interestingly , the 5′-phosphate is juxtaposed to the 3rd phosphate , and the Mg2+ ion coordinated by the carboxylate group of a C-terminal valine or leucine stabilizes these phosphates [12] , [20] . The distances between the two phosphate oxygen atoms and Mg2+ ion keep about 1 . 85 Å in all simulation time , indicating that the interactions between them are stable over time ( Table 2 ) , a result that is directly attributed to the effects of the metal ion which strengthens the protein-miRNA binding . This finding supports the idea that it is the 5′ phosphate that drags the whole miRNA strand and loads to the Mid domain to form a stable complex structure [40] . As compared to 5′-end , the 3′-end two nts show a low contribution ( ∼2% ) to the total energy , which supports the experimental conclusion that the PAZ-RNA binding potential is weak [12] . For the binding of mRNA to Ago-miRNA complex , from the analysis of ΔGbind-mR 12 most influential amino acids , i . e . , Arg9 , Lys97 , Arg102 , Arg440 , Arg543 , Arg569 , Lys570 , Lys594 , Lys611 , Phe640 , Arg654 and Lys657 were observed . These residues contribute 43 . 2% to the total ΔGbind-mR , with respective ΔGbind contributions >1 . 5% . Arg9 , Lys97 and Arg102 located in the PAZ domain are responsible for binding nucleotides 13–14 , and the rest ones responsible for binding nucleotides 10–12 of the target RNA ( Figure 9 and supporting Dataset S2 ) . Once again and also interestingly , we find that all these residues are also basic arginine ( 6/12 ) or even more basic lysine ( 5/12 ) , except one Phe640 ( 1/12 ) , leading to the conclusion that the mRNA-binding-channel in Ago is also of basic feature ( Figure 11 ) . Figure 10E shows the profile of contributions from each base of the miRNA in the ternary Ago-miRNA-mRNA to ΔGbind-mR . The 2–8 seed segments ( in dotted line square ) in miRNA totally contribute 7 . 1% to ΔGbind-mR , among which the 2nd , 3rd and 6th pairs ( G∶C ) are the most important yet the 4th one ( A∶U ) is the least one , indicating the significance of G∶C base-pairs in the seed region . The contributions from other non-seed bases are sharply diminished , which , on the other hand , demonstrates the importance of the seed region for miRNA-RNA binding [35] . The experiment indicates that extensive 3′ pairing of up to 17 nucleotides in the absence of the minimal 5′ element ( seed ) is not sufficient to confer regulation . And it also supports previous suggestion that the mismatch at positions 1 , 9 , or 10 did not affect the target gene function , but any mismatch in positions 2 to 8 could reduce the magnitude of target regulation [28] . In summary , our data are consistent with a picture whereby the seed region is the primary interaction site , whose relatively short length renders it sensitive to single mismatches . As for mRNA , except for the seed segment , U10′ plays the most important role in the target RNA binding , which contributes 12 . 0% to the ΔGbind-mR ( Figure 10F ) . The significance of U10′ is revealed by the base-independent strong electronic attraction between its backbone phosphate oxygen and Mg2+ ion ( Mg 679 ) in the PIWI domain ( Figure 8B ) . Overall , above observations provide compelling evidence that miRNAs recognize their targets mainly through limited base-pairing interactions between the 5′-end of miRNA and complementary sequences in the 3′ untranslated regions of the target mRNAs [41] , [42] , as well as through the interactions from amino acid residues in PIWI domain and the structural Mg2+ ion [20] . The energy relative contribution analysis for each amino acid and nucleotide was carried out based on the Generalized-Born decomposition approach , which is a major limitation of this work . Although in our work and many other cases , the net binding free energies are strongly correlated in the GB and PB models , a recent important work also found that the correlations of individual group contributions are highly variable that in some cases , strong correlation is seen , while in others , there is essentially none [43] .
The present MD simulations and thermodynamic analysis provide a new insight into the mechanism of the recognition of Argonaute to miRNA and mRNA at the atomic level . Our findings are summarized as follows: Our results provide an improved understanding of above key parameters that define how the guide RNA binds to Ago protein and their target genes . We anticipate that these will be of use in understanding known miRNA–target relationships as well as in improving methods to predict the miRNA targets .
As RNA always tends to decay in situ , it is difficult to determine the 3D structure of Ago/RNA complex , much less the structure of Ago/miRNA/mRNA complex . To date , Wang et al . have determined the X-ray crystal structures of three argonaute silencing complexes with a seed-containing guide DNA and target RNA duplex ( pdb codes: 3F73 [20] , 3HM9 [21] and 3DLB [31] ) , in which DNA-RNA hybrid was used mimicing the miRNA-mRNA complex . In addition , a NMR structure of a let-7:lin-41 complex from C . elegans was solved for the miRNA:mRNA complex ( pdb code: 2JXV [50] ) . In this work , in order to overcome the difficulty of unavailability of miRNA/mRNA/Ago complexes , several modeling and simulation methods were integrated to build the Ago complex structures: 1 ) A specific miRNA , i . e . , miR-7 was selected due to its validated importance for target regulation [28]; The starting B-RNA miR-7 ( 5′-UGGAAGACUA UGAUUUUGUUGU ) and its target mRNA ( 5′- ACAGCAGAAUCAAUAGUCUUCCG ) [28] structures for MD simulations were built using Bioploymer module in Sybyl 6 . 9 ( Tripos Inc . St . Louis , MO ) [51] , according to the basic features provided by the RNA-target complexes ( 2JXV . pdb [49] and 3F73 . pdb [20] ) . Based on the X-ray structure of the Ago silencing complexes ( 3F73 [20] , resolution 3 . 0 Å , 3HM9 [21] , 3 . 3 Å , 3DLB [31] , 2 . 7 Å ) , the 3D structural model of Ago was constructed . The side chains with missing coordinates were reconstructed using the fragment library of the Biopolymer module . The bound DNA-RNA fragments were removed from the original structure . Residues around the DNA-binding tunnel within 5 Å , which distance is large enough to include the binding site , were extracted to comprise the binding pocket for docking . Using the protein-protein docking and molecular superposition software Hex 4 . 2 ( http://www . csd . abdn . ac . uk/hex/ ) [52] , and the flexible docking program AutoDock 4 ( http://autodock . scripps . edu/ ) [53] , the possibly best binding pose ( orientation and conformation ) of RNA binding to Ago was searched . During the docking process , following steps were employed: The binary structure was obtained by removing the mRNA from the ternary system . Subsequently , the structures were treated with a rigid-body energy minimization in Hex to further optimize the pose , followed by molecular dynamics simulations . All molecular dynamics simulations were carried out using the GROMACS 4 . 0 package [55] on a 2 . 2 GHz 28 Intel Xeon workstation with 4×11 Gb of RAM with the AMBER99 force-field [56] . For MD simulations , three models ( free Ago , Ago-miRNA , and Ago-miRNA-mRNA complexes ) were solvated with the TIP3P water model [57] . The V-rescale thermostat [58] was applied using a coupling time of 0 . 1 ps to maintain the systems at a constant temperature of 300K , with pressure maintained by coupling to a reference pressure of 1 bar , and values of the isothermal compressibility set to 4 . 5×10−5 bar−1 for water simulations . Periodic boundary conditions were employed and Particle Mesh Ewald [59] was used for the long-range electrostatic interactions . Van der Waals and Coulomb interactions were truncated at 1 . 4 and 1 . 0 nm , respectively . All bond lengths including hydrogen atoms were constrained by the LINCS algorithm [60] . For each system , the simulation cell was a cubic periodic box with a side length of 106 Å , and the minimum distance between the protein and the box wall was set to be larger than 10 Å so that the protein does not directly interact with its own periodic image given the cutoff in every system . Numerical integration of the equations of motion used a time step of 2 fs , with non-bonded pair list updated every 10 steps and conformations stored every 2 ps for analysis . To neutralize the total charge and bring the tonicity of the solvent to physiological levels of 0 . 15 M , 107 Na+ and 109 Cl− , 107 Na+ and 91 Cl− , and 108 Na+ and 70 Cl− ions were placed randomly in the boxes of free Ago , Ago-miRNA and Ago-miRNA-mRNA complexes , respectively . In the end , the simulation system for free Ago is totally composed of 117492 atoms , with 10737 protein atoms and 35513 solvent molecules embedded in the cubic box . While the binary system ( Ago-miRNA ) contains a total of 117468 atoms , including 10737 protein atoms , 735 RNA atoms and 35266 solvent atoms . And the ternary system ( Ago-miRNA-mRNA ) is consisted of 12209 solution atoms and 35409 solvent atoms . After initial configuration construction , a standard equilibration protocol was performed for molecular simulations . The system was subjected to energy minimization for 5000 steps by steepest descent and for 15000 steps by conjugate gradient to avoid close atomic contacts , followed by slow constant volume heating to 300 K over 100 ps using 2 . 4 kcal/mol/Å2 harmonic restraints . These restraints were slowly reduced to zero during a series of energy minimizations and 50 ps equilibration steps at constant temperature ( 300 K ) and pressure ( 1 bar ) with a 0 . 2 ps coupling constant for both parameters . The final equilibration step was a 100 ps constant volume run . The production stage consisted of a total of 20 ns at constant temperature of 300 K for all the three systems , respectively . In order to identify the most significant fluctuation modes of the protein and RNAs , principal component analysis was carried out essentially as described by Wlodek et al [61] . The positional covariance matrix C was calculated from the equilibrated portion of each trajectory . For M snapshots of an N atom system , C is a 3N×3N matrix: ( 1 ) The eigenvectors of the covariance matrix , V , together with their corresponding eigenvalues , λ , were obtained by diagonalizing the covariance matrix C , i . e . , . The ( orthonormal ) eigenvectors provide a 3N-dimensional representation of principal modes of structural variation , and the eigenvalue for a mode indicates the relative contribution that this mode has made to motion within the trajectory , as calculated by . In the rotated Cartesian coordinate , the largest eigenvalue λ captures the largest fraction of the relative mean square fluctuation ( RMSF ) , and the second largest λ captures the next largest fraction of the RMSF , etc . Projections of the trajectory ( r ( t ) ) on major eigenvectors ( Eq . ( 2 ) ) were analyzed for their time-dependent behavior and probability distributions [62] . ( 2 ) The domain motion analyses based on the PCA results were performed using the DYNDOM program [29] . The binding free energy [63] was calculated by Eq . ( 3 ) as follows: ( 3 ) where ΔGb is the binding free energy in solution , ΔGintele and ΔGintvdw are electrostatic and van der Waals interaction energies between a protein and its ligand , respectively , ΔGsol is the solvation energy , and −TΔS is the contribution of conformational entropy to the binding . ΔGintele and ΔGintvdw were computed using the same parameter set as used in the MD simulation , with no cutoff applied to the calculation . Solvation energy ΔGsol was calculated using the MM/PBSA scripts supplied with AMBER 10 . 0 [64] . The free energy estimation involves separately evaluating the free energy for the solute and solvent for a series of snapshots and then averaging the results . Ensembles of the structures ( 500 snapshots ) for the MM/PBSA calculation were obtained from 6 ns MD simulations of the solvated complex . The electrostatic component of the solvation was estimated with a Poisson-Boltzmann electrostatic continuum method using program Delphi II [65] . The dielectric boundary is the molecular surface defined by a 1 . 4 Å probe sphere and by spheres centered on each atom with radii taken from the PARSE parameter set [66] 1 . 4 Å , with a value of 2 . 0 Å for the phosphorus . An interior dielectric of unity was used ( matching the dielectric chosen when evaluating the solute electrostatic interactions to scale the charges [67] , [68] ) and the outside dielectric was set to 80 . The nonpolar solvation energy was calculated from the solvent-accessible surface area ( SASA ) , obtained with the Amber molsurf module using a probe radius of 1 . 4 Å [69] with a parametrization of ( 4 ) where γ = 0 . 00542 kcal/Å2 mol and β = 0 . 92 kcal/mol . Estimated by a normal-mode analysis of the vibration frequencies , the entropy was calculated with Amber nmode module . This normal mode calculation is extremely time-consuming for large systems , thus only residues within 9 Å of the mass center of the ligand ( including the ligand , but excluding water molecules ) were used for the normal mode calculation [70] . Using a distance-dependent dielectric constant of ε = 4r , the truncated systems were minimized until the root-mean-square of the elements of the gradient vector was less than 5×10−5 kcal mol−1 Å−1 , and then the entropies were calculated using classical statistical formula [71] . 100 snapshots ( every fifth snapshot of 500 collected snapshots ) were used to estimate the contribution of the entropies of association [70] . The ligand-residue interaction decomposition was also performed by decomposing the total binding free energy into the contribution from each individual residue by the MM/GBSA method [72] , since PB energy decomposition is rather time-consuming and computationally expensive . The binding interaction of each ligand-residue pair includes three terms ( the entropy term is omitted because of its relative small contributions ) : electrostatic contribution ( ΔGele ) , van der Waals contribution ( ΔGvdw ) and solvation contribution ( ΔGsol ) . All energy components were calculated using 100 snapshots from 500 collected snapshots . | One of the biggest surprises at the beginning of the ‘post-genome era’ was the discovery of numerous genes encoding microRNAs . The number of microRNA genes is estimated to be nearly 1% of that of protein-coding genes , which were found in genomes of such diverse organisms as Caenorhabditis elegans , Drosophila melanogaster , Arabidopsis thaliana , and Homo sapiens . Their products , tiny RNAs ( miRNAs and siRNAs ) , are thought to bind to Argonaute ( Ago ) proteins and form effector complexes to direct mRNA cleavage or repress translation of complementary RNAs , during development , organogenesis , and very likely during many other processes . The cellular interactions between the miRNAs and their target RNAs associating with Ago are only beginning to be revealed , and details of this interaction mechanism at molecular level are still poorly understood . In this article we propose the possible mechanisms of miRNA-target interaction with special emphasis on their structural dynamic and thermodynamic aspects . The results of our model suggest the chemical and physical factors and effects that may be responsible for the miRNA-Ago assembly and miRNA-target recognition . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"computational",
"biology"
] | 2010 | Mechanism of MicroRNA-Target Interaction: Molecular Dynamics Simulations and Thermodynamics Analysis |
Let-7 miRNAs comprise one of the largest and most highly expressed family of miRNAs among vertebrates , and is critical for promoting differentiation , regulating metabolism , inhibiting cellular proliferation , and repressing carcinogenesis in a variety of tissues . The large size of the Let-7 family of miRNAs has complicated the development of mutant animal models . Here we describe the comprehensive repression of all Let-7 miRNAs in the intestinal epithelium via low-level tissue-specific expression of the Lin28b RNA-binding protein and a conditional knockout of the MirLet7c-2/Mirlet7b locus . This ablation of Let-7 triggers the development of intestinal adenocarcinomas concomitant with reduced survival . Analysis of both mouse and human intestinal cancer specimens reveals that stem cell markers were significantly associated with loss of Let-7 miRNA expression , and that a number of Let-7 targets were elevated , including Hmga1 and Hmga2 . Functional studies in 3-D enteroids revealed that Hmga2 is necessary and sufficient to mediate many characteristics of Let-7 depletion , namely accelerating cell cycle progression and enhancing a stem cell phenotype . In addition , inactivation of a single Hmga2 allele in the mouse intestine epithelium significantly represses tumorigenesis driven by Lin28b . In aggregate , we conclude that Let-7 depletion drives a stem cell phenotype and the development of intestinal cancer , primarily via Hmga2 .
Micro-RNAs ( miRNAs ) are critical for tumor suppression , which is most notably revealed following genetic manipulation of Dicer1 , an enzyme needed for miRNA processing , in which haplo-insufficiency of Dicer1 and global reduction of miRNA levels significantly accelerates tumorigenesis [1 , 2] . Let-7 miRNAs comprise one of the largest and most highly expressed families of miRNAs , possessing potent anti-carcinogenic properties in a variety of tissues [3] . This activity is likely mediated via Let-7 repression of a multitude of onco-fetal mRNAs and other pro-proliferative and/or pro-metastatic targets , such as HMGA2 , IGF2BP1 , IGF2BP2 , and NR6A1 [4–6] . Let-7 biogenesis is tightly regulated , revealed by the discovery of several proteins that regulate processing by DGCR8/DROSHA in the nucleus , and by DICER1 cleavage in the cytoplasm . Most notable are LIN28A and LIN28B , which are RNA-binding proteins that directly bind to and block the processing of Let-7 mRNAs [7 , 8] . LIN28A works in concert with TRIM25 and TUT4 to mediate terminal uridylation and subsequent degradation of immature precursor-Let-7 ( pre-Let-7 ) miRNA molecules [9–11] . LIN28B appears to act by sequestering primary-Let-7 ( pri-Let-7 ) miRNAs within the nucleolus to prohibit processing by DGCR8 and DROSHA [9] . The critical nature of maintaining sufficient levels of mature Let-7 miRNAs is reflected in the large number of studies that have found LIN28A or LIN28B up-regulated in human cancers , with expression often associated with an aggressive disease phenotype and/or predictive of poor outcomes [12–15] . LIN28B appears somewhat more frequently up-regulated than LIN28A in cancer , and may reflect the greater expression potential of LIN28B in adult tissues: LIN28B exhibits a later temporal pattern of expression in adult tissues such as the intestine , plays a greater role in post-natal growth , and can be re-activated by inflammation and NF-kappa-B [16–19] . In efforts of The Cancer Genome Atlas ( TCGA ) research consortia to define miRNA-mRNA associations across multiple different cancers ( i . e . the pan-cancer initiative ) , the LIN28B:Let-7b interaction was identified as one of the most significant relationships discovered across nine different human malignancies [20] . The tight functional interplay between LIN28 proteins and Let-7 is delineated clearly , on biochemical and biological levels . However , Let-7 action appears dependent on the particular mRNA targets affected , although Let-7 represses de-differentiation in multiple contexts . For example , Let-7 regulates insulin-PI3K-mTOR signaling in muscle by inhibiting expression of INSR , IGF1R , and IRS2 [21] , yet can also inhibit mTORC1 without affecting insulin-PI3K signaling [22] , whereas we have observed no effects on insulin-PI3K-mTOR signaling following depletion of Let-7 miRNAs in the small intestine [18] . Micro-RNAs have many targets , including both coding and non-coding mRNAs , and to address the functional impact of these miRNAs , one must dissect the cascade of integrated signals that ensue following alterations of a miRNA . Many studies have focused on RAS and MYC as cancer-relevant Let-7 targets , although recent high-throughput sequencing ( mRNA-seq , miRNA-seq , and CLIP-seq ) and meta-analyses indicate that these mRNA targets are not frequently regulated by Let-7 , especially in the context of cancer [5 , 6 , 20 , 23] . Onco-fetal Let-7 targets such as HMGA2 and IGF2BP1-3 appear to be more frequently up-regulated in multiple contexts , across multiple tissues , and in association with somatic stem cell potential [4 , 5 , 20 , 24–29] . We have demonstrated that LIN28B is a potent driver of colorectal cancer ( CRC ) progression , cellular invasion , and in mouse models , a regulator of intestinal growth and tumorigenesis [15 , 18 , 30] . The exploration of Let-7-dependence through genetic manipulation in mouse models is currently untenable due to the large number of miRNA clusters , with 12 Let-7 genes located at 8 separate clusters on 7 different chromosomes . To circumvent this obstacle and elucidate the mechanistic roles of Let-7 miRNAs in intestinal tumorigenesis in a genetic mouse model we have combined a Vil-Lin28bLow ( Lin28bLo ) transgene with intestinal deletion of the MirLet7c-2/Mirlet7b bi-cistronic cluster ( Let-7IEC-KO ) to achieve robust repression of all Let-7 miRNAs expressed in the intestinal epithelium . Concurrent deletion of the MirLet7c-2/Mirlet7b bi-cistronic cluster is necessary as Lin28b is unable to effectively target and inhibit processing of these specific Let-7 miRNAs [18] . These Lin28bLo/Let-7IEC-KO mice develop intestinal polyps with 100% penetrance and develop adenocarcinomas in the majority of animals , coincident with reduced survival . Examination of Let-7 targets in these tumors and in tumoroid cultures suggest that HMGA2 is likely playing a major role in driving carcinogenesis following Let-7 depletion , a novel in vivo finding . Furthermore , we find that tumorigenesis and a stem cell signature are driven by Let-7 depletion in mouse and human intestinal tumors , in which HMGA2 appears to play a functional role in reinforcing .
Vil-Lin28bLow mice and Let7IEC-KO mice were generated and described previously [18] . To generate compound mutant animals we used a low-expressing transgenic line ( Lin28bLo ) , in which we could not detect measureable changes in either protein or mRNA levels of Let-7-independent Lin28b targets [18] . These compound Lin28bLo/Let7IEC-KO mice , exhibit depletion of all Let-7 miRNAs specifically in intestinal epithelial cells ( IEC ) achieved through deletion of the MirLet7c-2/MirLet7b locus and repression of all other Let-7 miRNAs through inhibition by Lin28b [18] ( and Fig 1A ) . Lin28bLo/Let7IEC-KO mice thrived initially , with normal behavior and weight gain , but displayed significantly increased mortality and morbidity starting around 6 months of age , whereas neither Vil-Lin28bLo nor Let7IEC-KO age-matched mice exhibited any overt phenotype ( Fig 1B ) . Surviving Lin28bLo/Let7IEC-KO were sacrificed between 10 and 14 months of age and exhibited a significant incidence of adenomas and adenocarcinomas , restricted to the small intestine , with an average of 2 . 86 lesions per mouse and 100% penetrance ( S1 Table and Fig 1C , 1D and 1E ) . Six of seven Lin28bLo/Let7IEC-KO mice developed invasive adenocarcinoma ( S1 Table and Fig 1C , 1D and 1E ) . Tumors from mice also displayed nuclear localization of β-catenin ( Fig 1F ) , indicative of constitutive activation of the Wnt signaling pathway . The severity of the Lin28bLo/Let7IEC-KO phenotype was substantially more dramatic than in Vil-Lin28bLo or Vil-Lin28bMed mice ( 18 ) . Vil-Lin28bMed mice express higher levels of Lin28b , have partially depleted Let-7 miRNAs and develop adenocarcinomas of the small intestine as do Lin28bLo/Let7IEC-KO mice but do not exhibit a phenotype as severe as Lin28bLo/Let7IEC-KO mice ( 18 ) . Let-7 targets were examined in small intestine crypts from Vil-Lin28b and Lin28bLo/Let7IEC-KO mice . RNA microarray expression analysis was previously performed on Vil-Lin28bMed total small intestine epithelia and we verified elevation of Hmga1 , Hmga2 , Igf2bp1 , Igf2bp2 , E2f5 , Acvr1c , Nr6a1 , Hif3a , Arid3a , Plagl2 , Trim6 , Ddx19a , and Mycn ( Fig 2A and [18] ) . We also observed significant elevation of mRNAs for these Let-7 targets in crypts from small intestine epithelia from Lin28bLo/Let7IEC-KO ( Fig 2B ) . Expression of all Let-7 targets also correlated significantly between Lin28bLo/Let7IEC-KO and Vil-Lin28bMed intestine crypts , with Hmga2 , Igf2bp2 , Hif3a , Arid3a , and E2f5 being the most highly induced targets in both models ( Fig 2C ) . All targets contained conserved Let-7 sites in the 3’UTR or coding sequence , except for Trim6 , for which only the mouse mRNA possesses Let-7 sites ( Fig 2D ) . In addition to our findings for HMGA2 , IGF2BP1 , and IGF2BP2 , there is experimental evidence that HMGA1 , E2F5 , and ARID3A are also direct targets of Let-7 [6 , 31 , 32] . To gain insight into the association of several Let-7 targets with tumorigenesis in vivo , we examined Hmga1 , Hmga2 , Arid3a , and Hif3a protein expression by immunostaining adenomas and adenocarcinomas , as well as adjacent normal tissue , from Lin28bLo/Let7IEC-KO mice . These targets exhibited little or modest up-regulation in normal small intestine epithelia of Lin28bLo/Let7IEC-KO mice , but dramatic increases in tumors ( Fig 3A–3J and S3A–S3H Fig ) . Pathological assessment of the staining pattern revealed that Hmga1 and Hmga2 staining was most intense in areas of invasive adenocarcinoma ( Fig 3G , 3H and 3I ) . We next examined Let-7 targets that might mediate programs of tumorigenesis in Lin28bLo/Let7IEC-KO mice in the context of tumors and cellular transformation . To model intestinal epithelial carcinogenesis we developed a 3-D culture model to examine only the epithelium and to select transformed tumor cells ( Fig 4A ) . Enteroids derived from CRC tumors appear to faithfully recapitulate the major expression signatures of un-manipulated whole tumors [33] . To pursue this , we micro-dissected and dissociated adenocarcinomas from Lin28bLo/Let7IEC-KO mice and cultured “tumoroids” from these lesions in medium supplemented with EGF , Noggin , and Rspo1 , as described previously for enteroid culture [34] . These tumoroid/enteroids ( T/E ) resembled normal small intestine enteroids ( Fig 4B ) and are likely a mixture of different cell types , but upon withdrawal of Noggin and Rspo1 , a small population of growth-factor independent cells expanded into tumoroid cysts ( TC ) ( Fig 4C ) , which likely possess cell-autonomous activation of Wnt signaling and Noggin-independent resistance to BMP signaling . Quantification by Taqman RT-PCR confirmed that Let-7 miRNAs are severely repressed in tumoroid/enteroids and transformed tumoroid cysts ( Fig 4D ) . Tumors and tumoroids , but not normal tissue from Lin28bLo/Let7IEC-KO mice , also exhibited up-regulation of Wnt target genes Axin2 , CD44 , and cMyc ( Fig 4E ) , suggesting spontaneous and constitutive activation of Wnt signaling . Analysis of Let-7 target mRNAs revealed two basic patterns of expression , with one group displaying expression highest in intact tumors or tumoroids/enteroids ( Fig 4F ) . The other group displayed increasing or plateauing expression , with higher levels in tumoroid/enteroids or tumoroid cysts ( Fig 4G ) . In this latter group we find known and suspected oncogenes , such as Hmga1 , Hmga2 , Igf2bp1 , Igf2bp2 , and Mycn . As Hmga2 appeared to exhibit pronounced up-regulation ( >200-fold ) in the tumoroid/enteroid and tumoroid cyst populations , and increased staining in invasive areas of adenocarcinomas ( Fig 3H ) , we evaluated Hmga2 co-localization with nuclear β-catenin in mouse tumors , to assay potential coincident activation of canonical Wnt signaling with nuclear Hmga2 . Immunostaining in both adenomas and adenocarcinomas from Lin28bLo/Let7IEC-KO mice revealed frequent and intense co-staining of Hmga2 with nuclear β-catenin ( Fig 4H–4K ) . This pattern of co-staining was not observed for Hmga1 , Arid3a , or Hif3a . To extrapolate relevance to human CRC from these mouse models , we examined expression data from human samples from The Cancer Genome Atlas ( TCGA ) [35] by querying for expression of Let-7 target mRNAs , with a focus on targets that exhibited significant up-regulation in either Vil-Lin28bMed or Lin28bLo/Let7IEC-KO mouse models ( namely , ARID3A , PLAGL2 , HMGA1 , HMGA2 , MYCN , IGF2BP1 , IGF2BP2 , and E2F5 ) . We examined a cohort of 416 CRC patients from a TCGA dataset and found that all transcripts except HIF3A mRNA were significantly elevated in cancer tissue compared to expression levels in normal tissues ( Fig 5A ) . IGF2BP1 expression in primary tumors was also associated with an increased likelihood of having nodal metastases ( Fig 5A ) . Levels of HMGA1 , HMGA2 , PLAGL2 , IGF2BP2 , E2F5 , and ARID3A transcripts were also inversely proportional to levels of Let-7 miRNA by examination of a cohort of 199 CRC patients from the TCGA Pan-Cancer analysis project [20] ( Fig 5B–5E and S1D–S1I Fig ) . Since Let-7a and Let-7b appear to be the most highly expressed Let-7 miRNAs in normal colonic epithelium , and are significantly depleted in CRC specimens [20 , 30] ( S1A , S1B and S1C Fig ) , we examined these miRNAs in a subset of colon cancer specimens . We also compared their expression with the crypt-base-columnar ( CBC ) stem cell markers EPHB2 , ASCL2 , and LGR5 , which are markers of stem cells in human intestine/colon and CRC , and are associated with aggressive CRC [36] . We found that Let-7a and Let-7b were significantly down-regulated in CRC specimens , while stem cell markers were significantly up-regulated ( Fig 5F and 5G ) . Let-7a and Let-7b levels were also correlated tightly , suggesting co-regulation ( Fig 5H ) , and were also inversely proportional to the expression of the stem cell markers EPHB2 and LGR5 ( Fig 5I ) . This suggests provocatively that Let-7a and Let-7b depletion may contribute to a stem cell phenotype in the intestine , and perhaps CRC . To further examine this relationship we evaluated small intestine stem cell markers in wild-type intestine , Lin28bLo/Let7IEC-KO intestine , and in tumors from Lin28bLo/Let7IEC-KO mice . In normal adjacent tissue we observed a trend toward increased expression of multiple stem cell markers in Lin28bLo/Let7IEC-KO small intestine ( Fig 5J ) . In contrast , tumors from Lin28bLo/Let7IEC-KO mice exhibited a pronounced up-regulation of all stem cell markers assayed , including Bmi1 , Lrig1 , Olfm4 , Ascl2 , Prom1 , Lgr5 , Msi1 , and Sox9 ( Fig 5J ) , perhaps suggesting an expansion of CBC and +4 stem cell-like compartments . While Let-7a and Let-7b depletion and increased expression of stem cell markers may appear to be a general feature of colon cancer , our discovery of a relationship between expression of Let-7 and stem cell markers suggests a functional connection . To examine a possible relationship between Let-7 target mRNAs and stem cell markers , we evaluated co-expression in mouse samples ( from Fig 5I ) and found that Hmga1 and Hmga2 had very high correlation with all of the markers we examined ( Fig 5K ) . Likewise , in human CRC samples the expression of HMGA2 directly correlates with LGR5 levels ( Fig 5L ) . In order to explore any disease relevance connecting HMGA1 and HMGA2 expression and tumor phenotype , we stained CRC tumor tissue arrays for these proteins and evaluated expression in relationship to various parameters including tumor stage , histopathologic characteristics , and disease outcomes . High-level HMGA1 expression predicted poor survival for patients with stage II tumors ( S2A Fig ) . HMGA1 staining was also more intense in stage II tumors ( S2C and S2D Fig ) and in tumors with perineural invasion ( S2E and S2F Fig ) . Interestingly , expression data from TCGA mRNA-seq studies [37] indicated that high-level expression of HMGA2 correlates inversely with survival ( S2B Fig ) . In tissue arrays HMGA2 expression was greater in non-mucinous tumors ( S2G and S2H Fig ) and in stage III tumors ( S2I and S2J Fig ) . In aggregate , these data suggest that HMGA1 and HMGA2 are expressed in non-overlapping tumor types , but are both associated with more aggressive phenotypes , and perhaps reduced patient survival . We next pursued 3-D culture and manipulation of intestinal organoids ( enteroids ) to explore the relationship between Let-7 targets and a stem cell phenotype . This method has facilitated the examination of stem cell phenotypes in the intestinal epithelium in multiple studies [34 , 38–44] . For these experiments we derived enteroids from Vil-Lin28bMed mice [18] . We have previously shown that crypt hyperplasia and Hmga2 expression is dependent on Let-7 depletion in crypts from Vil-Lin28bMed mice [18] . Enteroids derived from Vil-Lin28bMed mice exhibited enhanced colony forming potential of single cells ( Fig 6A , 6B and 6C ) . This is unlikely to be a feature secondary to enhanced stem cell potential conferred by increased association with Paneth cells , as described previously [34] , since this cell type is severely depleted following Let-7 repression [18] . To assay exogenous expression of Let-7 targets in enteroids , we used a lentivirus vector for transduction of wild-type mouse small intestine enteroids ( Fig 6D–6G ) . This vector system yields low ( Fig 6J ) or high-level ( Fig 6K ) expression , in a doxycycline-dependent manner . We generated stable enteroid lines for inducible expression of mouse Hmga2 , Igf2bp2 , E2F5 , Arid3a , or Hif3a and assayed colony forming potential and EdU incorporation . We focused on Hmga2 , rather than Hmga1 , as it is consistently up-regulated in non-malignant intestinal tissue from Vil-Lin28bMed and Lin28bLo/Let7IEC-KO and thus appears highly dependent on Let-7 [18] . For colony formation , only Hmga2 over-expression ( O/E ) exhibited a significant effect , with enhanced formation of new enteroids from singly plated cells ( Fig 6L , 6M and 6N ) , whereas Igf2bp2 , E2F5 , Arid3a , and Hif3a had no apparent effect ( Fig 6L and S4A Fig ) . Expression of Hmga2 , Arid3a , Hif3a , or Igf2bp2 via lentiviral vectors did not induce any change in stem cell markers ( S4B–S4K Fig ) . To determine if Hmga2 was necessary for the enhanced colony formation in Vil-Lin28b enteroids , we crossed Vil-Lin28bMed mice onto background in which one allele of Hmga2 is inactivated specifically in the intestine ( Vil-Cre+/Hmga2CK/+ ) [45] , and generated enteroids . We used Vil-Lin28bMed mice because their phenotype appears Let-7-dependent [18] and for simpler breeding . Effects on colony formation by Lin28b were greatly blunted by inactivation of a single Hmga2 allele ( Fig 6O ) . Hmga2 could also trigger increased EdU incorporation in intestinal enteroids , whereas Hif3a repressed it , suggesting opposing effects of Hmga2 and Hif3a on cellular proliferation ( Fig 6P and 6Q ) . Lentiviral-mediated expression and manipulation of the Hmga2 conditional allele were restricted to coding sequence only [45] . Perhaps consistent with its association with a stem cell phenotype , HMGA2 is also frequently co-expressed with the stem cell markers MSI1 and LGR5 in human CRC , and notably , more frequently than any of the other Let-7 targets evaluated here in this study ( Fig 5L and S2 Table ) . Lastly , to evaluate the role of Hmga2 in intestinal tumorigenesis in the context of Let-7 depletion we examined tumor burden in Vil-Lin28bMed and Vil-Lin28bMed/Hmga2+/IEC-KO mice . As mentioned earlier , Vil-Lin28bMed mice have a lower penetrance of intestinal tumorigenesis compared to Lin28bLo/Let7IEC-KO mice , with about 50% of animals developing tumors by 9 months of age ( S3 Table ) . Inactivation of one allele of Hmga2 in the intestinal epithelium significantly reduced disease penetrance and tumor burden in Vil-Lin28bMed/Hmga2+/IEC-KO mice ( S3 Table ) .
We have achieved comprehensive depletion of all Let-7 miRNAs in the intestinal epithelium and demonstrated the critical nature of their cumulative tumor-suppressive properties . These effects appear to be due to Let-7 , although LIN28B can bind mRNAs and modulate protein levels of targets in the intestinal epithelium [18] . However , this appears unlikely in Lin28bLo/Let7IEC-KO mice since LIN28B did not have any effect on RNA or protein levels of targets in the context of low-level expression in Vil-Lin28bLo mice [18] . While tumors from Lin28bLo/Let7IEC-KO mice appear to be more advanced than those from Vil-Lin28bMed mice [18] , surprisingly there is not a significant difference in tumor multiplicity . Nascent tumorigenesis beginning with aberrant crypt foci and/or microadenomas may occur spontaneously in our mouse model of Let-7 depletion , likely due to sporadic deregulation of Wnt signaling or potential spontaneous loss of other tumor suppressive mechanisms . Therefore , Let-7 may not have the “gatekeeper” function that is characteristic of tumor suppressors such as APC . Despite this , there is a link between LIN28B expression in human colon cancer samples and aggressive disease in early stages , which may reflect a role for LIN28B in early neoplastic growth [15] . Supporting this hypothesis is the documentation that LIN28 proteins and Let-7 miRNAs do indeed affect proliferation , migration , and invasion in cell culture models and xenografts of various malignancies [16 , 17 , 46–49] . However , the differences between Let-7 target mRNAs in each of these models can be quite disparate; e . g . KRAS has a larger effect on tumorigenesis than does HMGA2 in a non-small cell lung cancer model [49] , whereas HMGA2 appears to have a much larger role in other cancer models [28 , 50–53] , likely as a modifier of chromatin structure and gene expression [54–57] . As documented in developmental programs in C . elegans and in human cancers , Let-7 miRNAs repress a stem cell phenotype and tumor-initiating phenotype [3] , an association we observe here as well . The connection between HMGA2 and a stem cell phenotype in the intestinal epithelium is also intriguing . HMGA2 promotes somatic stem cell specification , with such roles in neural stem cells and hematopoietic stem cells [25–27] . In some contexts , HMGA2 can enhance Wnt signaling , a known driver of the crypt-base-columnar ( CBC ) intestinal epithelial stem cell phenotype [34 , 58 , 59] . This was observed in a mouse model of prostatic intraepithelial neoplasia , where overexpression of Hmga2 in cancer-associated fibroblasts enhances expression of the Wnt ligands Wnt2 , Wnt4 , and Wnt9b , concomitant with enhanced Wnt signaling and nuclear β-catenin in adjacent neoplastic epithelium [60] . Wnt signaling is required for the enhanced prostatic intraepithelial tumorigenesis induced by Hmga2 in this model [60] . However , we do not observe any effects on Wnt target genes or β-catenin localization in non-malignant Lin28bLo/Let7IEC-KO intestine tissue , suggesting that Wnt deregulation may be an independent event . A recent study that largely replicated our earlier work found that tumors triggered by transgenic LIN28B expression in the mouse small intestine frequently have mutations in Ctnnb1 ( β-catenin ) , but not Apc [61] . Although the level of induced LIN28B expression in this study is likely much higher than in Lin28bLo/Let7IEC-KO mice , and therefore different , we suspect that derangements of the Wnt pathway , e . g . in Ctnnb1 , are also occurring in tumors in Lin28bLo/Let7IEC-KO mice , as evidenced by frequent nuclear β-catenin . Alternatively , the co-localization of nuclear β-catenin with intense Hmga2 staining in mouse tumors ( Fig 6K–6N ) could reflect Wnt signaling enhancement of Hmga2 expression , a phenomenon observed in triple-negative breast cancer , a subtype that also tends to express high LIN28B levels [9 , 62] . Curiously , in our genetic manipulations of enteroids , exogenous Hmga2 does not affect expression of stem cell markers ( such as those assayed in Fig 6I ) in transduced intestinal enteroids ( S4 Fig ) . Alternatively , increased Hmga2 expression may enhance survival of stem cells or facilitate the recruitment of a facultative population ( such as the quiescent “+4” secretory progenitor stem cell ) and entry into the cell cycle . Or , Hmga2 may synergize with deregulated Wnt signaling in the promotion of a stem cell phenotype , which could account for the dramatic up-regulation of stem cell markers we see in tumors from Lin28bLo/Let7IEC-KO mice . Others have also reported that HMGA2 expression is predictive of aggressive disease and poor outcomes in CRC [63] , as similarly found in other cancers [50] . While HMGA2 is playing a key role , it is likely that the effects of Let-7 on an intestinal stem cell phenotype and epithelial tumorigenesis are dependent on the collective and/or cooperative role of multiple Let-7 targets . Not uncommonly , additive roles of target genes are uncovered in the genetic dissection of a single pathway , such as that seen for PDGF-receptor signaling and the collective biological contribution of multiple target genes dependent on Pdgfra and Pdgfrb [64] . However , it is challenging to dissect the combinatorial relationships among a dozen candidate targets , especially in mouse models . An oncogenic function of HMGA1 and IGF2BP1 has been reported in other cancers , including colon cancer , with evidence that both factors enhance tumorigenesis [65 , 66] . Dissecting the interaction and possible cooperation of Let-7 target mRNAs is critical for designing strategies to ameliorate the loss of Let-7 in human cancers via combinatorial targeted therapies against multiple oncogenes .
Mouse studies were approved by the University of Pennsylvania Animal Care and Use Committee , protocol #802791 . Vil-Lin28bLo , Vil-Lin28bMed , and Let7IEC-KO mice were described previously [18] , and were maintained via backcrosses to C57BL/6J . Vil-Lin28bMed mice express Lin28b protein approximately 2-fold higher than Vil-Lin28bLo mice . To obtain Lin28bLo/Let7IEC-KO mice , VilCre+/Let7lox/lox mice were mated with Vil-Lin28bLo/Let7lox/lox mice to get Vil-Lin28bLo/VilCre+/Let7lox/lox , and all other possible genotypes . Let7lox/lox mice were considered wild-type and possess all Let-7 miRNAs at levels insignificantly different from wild-type mice [18] . Conditional null Hmga2Ck mice were described previously [45] . Mice were sacrificed at 12 weeks or between 10 and 14 months of age for dissection , isolation of tissues for histology and immunohistochemistry , and isolation of intestinal epithelia . Pathological criteria for mouse intestinal tumors were used as previously defined [67] . Intestinal adenomas are exophytic growths without evidence of invasion characterized by enlarged variably hyperchromatic nuclei with altered glandular architecture , including enlarged crypts and budding , irregular glands . For purposes of defining an adenocarcinoma , tumor invasion through the lamina propria into the muscularis mucosa and eventually beyond must clearly be seen . For mRNA expression analysis of mouse tissue , whole jejunum , total intestinal epithelium , or crypt epithelium was isolated for homogenization in Trizol ( Life Technologies ) . Total epithelium or crypts were isolated as described previously [18] . Total RNA ( 2–5 μg ) was used for cDNA synthesis with oligo dT primers and Superscript III RT ( Life Technologies ) according to the manufacturer instructions . QPCR was performed using Taqman technology or Sybr green using the TaqMan Fast Universal PCR Master Mix ( 2X ) , no AmpErase UNG ( Life Technologies ) or the Power SYBR Green PCR Master Mix ( Life Technologies ) . Expression levels of queried mRNAs were normalized to β-actin ( Actb ) and Hprt or Gapdh mRNA levels . Let-7 miRNAs were quantified using Taqman Q-RT-PCR kits ( Life Technologies ) , according to the manufacturers instructions and normalized to U6 and SNO135 small RNA levels . Primers and Taqman probes are listed in S3 Table . Human colon cancer tumor specimens , along with adjacent matched non-malignant tissue , were obtained from the Siteman Cancer Center Tissue Procurement Core as fresh frozen sections . Twenty samples ( 11 pairs ) were obtained and total RNA was isolated following homogenization with Trizol ( Life Technologies ) . For qualitative evaluation of RNA integrity 2 μg of total RNA was electrophoresed on a 1% agarose gel . For evaluation of mRNAs , 1 μg of total RNA was used for reverse transcriptase using the iScript reverse transcriptase kit ( BioRad ) , while miRNAs were quantified using Taqman Q-RT-PCR kits ( Life Technologies ) , according to the manufacturers instructions . Levels of mRNAs were assayed using standard primer pairs and SsoFast EvaGreen Supermix ( Biorad ) and normalized to cyclophilin-A ( PPIA ) and β-2 microglobulin ( B2M ) . Let-7a and Let-7b miRNAs were normalized to U6 and RNU6B RNAs . Crypts were isolated as described previously [18] and cultured in EGF , Noggin , and Rspo1 ( ENR ) medium [34] . Before plating , crypts were counted and re-suspended in a mixture of 80% Matrigel ( BD Biosciences ) and 20% ENR at a concentration of 20 crypts per μl . For initial plating and the first three days of culture , crypts were grown in the presence of 10 μM Rho kinase ( ROCK ) inhibitor ( Y27632 ) . Medium was then changed every 3 days with fresh ENR medium . For lentiviral transduction the pTRIPz vector was modified for expression of mouse Hmga2 ( NM_010441 . 2 ) , E2f5 ( NM_007892 . 2 ) , Igf2bp2 ( NM_183029 . 2 ) , Arid3a ( NM_001288625 . 1 ) or Hif3a ( NM_016868 . 3 ) by cloning the open reading frames from these cDNAs between the AgeI and MluI sites within pTRIPz . Enteroids were transduced with lentiviral particles as described previously and selected with 2 μg/ml puromycin [68] . Enteroids were mechanically dissociated by pipetting up and down in 4 ml basal medium and centrifuged at 100 x g for 2 min . Enteroids were then re-suspended in 0 . 5 ml TrypLE Express ( Life Technologies ) containing 250 U/ml DNase I ( 1:200 ) and 10 μM ROCK inhibitor . Enteroids were incubated 5 minutes at 37°C with periodic vortexing every 60 sec . To this we added one volume basal medium with 5% FBS ( with DNase and ROCKi ) and spun 5 min at 200 x g . Cells were re-suspended in 0 . 5 ml pre-warmed basal medium with 50 U/ml DNase I and 10 μM ROCK inhibitor and incubated 5 minutes at room temperature with periodic vortexing . Single cells were then plated in triplicate at a concentration of 2500 cells per cm2 in 80% Matrigel , 20% ENR , and over-layed with ENR medium plus 10 μM ROCK inhibitor . Colonies of growing , budding enteroids were counted 5–7 days after plating . For assaying 5-ethynyl-2´-deoxyuridine ( EdU ) incorporation , enteroids were given fresh new medium containing 10 μM EdU and incubated for 2 hours . Enteroids were then isolated as performed above to obtain a single cell suspension , then fixed for Click-iT labeling and flow cytometry using the Click-iT Plus EdU Alexa Fluor 488 Flow Cytometry Assay Kit ( Life Technologies ) . Fixation and labeling was carried out according to the manufacturer instructions . Mouse intestinal tumors were isolated for culture by micro-dissecting tumor tissue away from normal adjacent mucosa using a dissecting microscope . Two pieces of tumor were processed for RNA isolation and histology and immunohistochemistry while a third piece was dissociated for culture . For tumoroid culture , tumors were placed into HBSS containing 10 mM EDTA , 1 mM N-acetyl-cysteine ( NAC ) , and 10 μM ROCK inhibitor ( Y27632 ) and incubated at 37°C with periodic vortexing for approximately 5 to 10 minutes until the epithelium began to detach . Isolated epithelium was then washed three times with sterile basal medium and plated in culture as done above for enteroids in ENR medium . After 1–2 weeks of continuous culture and 1–2 passages , tumoroids were placed into medium lacking Noggin and Rspo1 . While most enteroids died , small rare surviving colonies could be observed after 3–5 days of culture . These tumoroid cysts were maintained in medium lacking Noggin and Rspo1 . Paraffin-embedded enteroids , intestinal tissue , and tissue microarrays were incubated at 56°C prior to de-waxing and rehydration . Antigens were retrieved by boiling sections in 10 mM citric acid , pH 6 . 0 , for 2 hrs . Samples were blocked in 1% BSA , 0 . 3% Triton-X-100 , and 10% normal goat serum for 1 hr . Endogenous peroxidases were quenched in 3% hydrogen peroxide for 5 minutes . In conjunction with biotin-conjugated secondary antibodies ( Jackson ImmunoResearch , diluted 1:200 ) stains were developed with the VECTASTAIN Elite ABC Kit ( Vector Laboratories , cat# PK-6100 ) and the DAB Peroxidase ( HRP ) Substrate Kit ( Vector Laboratories , cat# SK-4100 ) . Tissues were dehydrated and cover-slipped with Cytoseal ( Thermo Scientific , cat# 8310–4 ) . Primary antibodies used for IHC were anti-ARID3A ( 1:100 , ProteinTech , Chicago IL , cat# 14068-1-AP ) , anti-β-catenin [D10A8] XP Rabbit mAb ( 1:100 , Cell Signaling , Danvers MA , cat# 8480 ) , rabbit anti-HIF3A antibody ( 1:200 , Sigma-Aldrich , St . Louis MO , cat# SAB2900410 ) , anti-HMGA1 antibody [EPR7839] ( 1:250 , Abcam , Cambridge MA , cat# ab129153 ) , and anti-HMGA2 [D1A7] rabbit mAb ( 1:400 , Cell Signaling , Danvers MA , cat# 8179 ) . We examined gene expression in CRC specimens from a cohort of 416 CRC patients from a TCGA dataset using the cancer genome browser at UCSC ( https://genome-cancer . ucsc . edu/proj/site/hgHeatmap/ ) ( Cline MS 2013; Lopez-Bigas N 2013; Goldman M 2012; Sanborn JZ 2011; Vaske CJ 2010; Zhu J 2009 ) . For examination of Let-7 miRNA expression and expression relative to candidate target genes we examined a cohort of 199 CRC patients from the TCGA Pan-Cancer analysis project visualized using the starbase miRNA CLIP-seq portal ( http://starbase . sysu . edu . cn/ ) ( Li JH et al . , Nucleic Acids Res . 2014; Yang JH et al . , Nucleic Acids Res . 2011 ) . | Cancer develops following multiple genetic mutations ( i . e . in tumor suppressors and oncogenes ) , and mutations that cooperate or synergize are often advantageous to cancer cell growth . To study how multiple genes might cooperate , it is usually informative to generate candidate mutations in cells or in mice . Large gene families , such as the Let-7 family , are difficult to silence or mutate because of the large amount of redundancy that exists between similar copies of the same gene; the mutation of one will often be masked or compensated by the continued function of others . In the mouse intestine we have achieved comprehensive depletion of all Let-7 miRNAs in this large multi-genic family through use of an inhibitory protein , called LIN28B , that specifically represses Let-7 , and genetic inactivation of another gene cluster called MirLet7c-2/Mirlet7b . Mice with this comprehensive depletion of Let-7 develop intestinal cancers that resemble human colon cancers . Our further analysis identified another gene , HMGA2 , downstream of this pathway that is critical to this outcome . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Let-7 Represses Carcinogenesis and a Stem Cell Phenotype in the Intestine via Regulation of Hmga2 |
Rhodopsin has been used as a prototype system to investigate G protein-coupled receptor ( GPCR ) internalization and endocytic sorting mechanisms . Failure of rhodopsin recycling upon light activation results in various degenerative retinal diseases . Accumulation of internalized rhodopsin in late endosomes and the impairment of its lysosomal degradation are associated with unregulated cell death that occurs in dystrophies . However , the molecular basis of rhodopsin accumulation remains elusive . We found that the novel norpAP24 suppressor , diehard4 , is responsible for the inability of endo-lysosomal rhodopsin trafficking and retinal degeneration in Drosophila models of retinal dystrophies . We found that diehard4 encodes Osiris 21 . Loss of its function suppresses retinal degeneration in norpAP24 , rdgC306 , and trp1 , but not in rdgB2 , suggesting a common cause of photoreceptor death . In addition , the loss of Osiris 21 function shifts the membrane balance between late endosomes and lysosomes as evidenced by smaller late endosomes and the proliferation of lysosomal compartments , thus facilitating the degradation of endocytosed rhodopsin . Our results demonstrate the existence of negative regulation in vesicular traffic between endosomes and lysosomes . We anticipate that the identification of additional components and an in-depth description of this specific molecular machinery will aid in therapeutic interventions of various retinal dystrophies and GPCR-related human diseases .
Retinitis pigmentosa is the most common form of retinal degeneration and the major cause of human blindness [1] . Most degenerative retinal dystrophies are caused by various genetic mutations . Malfunctioning of phototransduction is the predominant cause of retinal dystrophies , in that most genes involved in the functioning and regulation of the phototransduction cascade are directly or indirectly related to retinal degeneration syndromes [2] , [3] . Therefore , it is not surprising that rhodopsin-1 , the major visual pigment , was the first molecule identified as a target for such mutations [4] , [5] . Drosophila norpA ( phospholipase C , PLC ) acts as a central effector molecule in phototransduction [6] . It has been used as an invertebrate model for studying molecular mechanisms of retinal degeneration caused by malfunctioning of the phototransduction cascade [7] . Interestingly , cGMP phosphodiesterase , which relays the signal from G-proteins in vertebrate phototransduction , is also known to trigger retinal degeneration in mouse models [8]–[10] . The loss of norpA function essentially shuts down the phototransduction cascade , resulting in a failure to raise intracellular Ca2+ levels through light-sensitive channels . Thus , Ca2+-dependent enzymes required for rhodopsin recycling cannot be activated , resulting in the formation of the stable rhodopsin-arrestin complex [11]–[14] . It has been reported that excessive endocytosis followed by the formation of stable rhodopsin-arrestin complexes and accumulation of internalized rhodopsin in late endosomes trigger apoptosis in norpA mutant photoreceptor cells [12] . The “granule group” genes in Drosophila have been known for their vital role in lysosomal biogenesis and functioning [15] , [16] . A previous study found that the functional loss of the “granule group” genes resulted in rhodopsin accumulation in the Rab7-positive late endosomes and triggered retinal degeneration in norpA mutant photoreceptor cells [12] , [17] . Therefore , the accumulation of internalized rhodopsin in late endosomes and impaired endo-lysosomal trafficking clearly causes retinal degeneration in both the norpA and the “granule group” mutant photoreceptors . However , the molecular basis of this pathologic accumulation remains unknown . The role of excessive endocytosis of light-activated rhodopsin on saturating the capacity of the trafficking machinery for the endo-lysosomal progression , resulting in the accumulation of endocytosed rhodopsin in the late endosomes remains controversial . Alternatively , previously unknown regulatory mechanisms prevent endocytosed rhodopsin from further movement toward lysosome . A growing number of evidences support the fact that the eukaryotic cell utilizes active regulatory mechanisms in monitoring and maintaining the intracellular membrane balance of the endo-lysosomal system [18]–[21] . Therefore , it is imperative to identify genetic components underlying rhodopsin accumulation and present epistatic evidences that possibly override the endo-lysosomal blockage in phototransduction mutants . Triplo-lethal ( Tpl ) locus , cytologically defined as the 83D4-E2 region in chromosome 3 in Drosophila , was identified as a sole locus responsible for both triplo-lethality and haplo-lethalith in segmental aneuploids [22] . Point mutations responsible for the Tpl phenotype remain unidentified [23] , although the Ell product , a transcription elongation factor , was found to be a suppressor of the Tpl phenotype [24] . Therefore , it is proposed that this phenotype is caused by a gene cluster that shows at least partial redundancy and its dosage is critical to its function [25] . The Osiris gene family was identified in an effort to explain the Tpl phenotype as an effect of a gene cluster . This is a large conserved family , with most genes ( 20 of 23 ) located within the Tpl locus [26] . Although the cellular function of the Osiris family of proteins is currently unknown , all members share characteristic features , including endoplasmic reticulum signal sequences , a pair of cysteine residues near the amino terminus , a putative transmembrane domain , an AQXLAY motif , and a number of endocytic signaling motifs such as YXXØ [26] , [27] . Previously , we screened for norpAP24 suppressors by random mutagenesis . The screening had the advantage of the yeast site-specific recombination FLP-FRT system and could identify both essential and nonessential genes [28] . Here we report that the novel norpAP24 suppressor , diehard4 ( die4 ) , is responsible for the inability of endo-lysosomal rhodopsin trafficking and retinal degeneration in norpAP24 mutants . We found that die4 encodes Osiris 21 ( Osi21 ) . A loss of function of Osi21 suppresses retinal degeneration in various phototransduction mutants . In addition , the loss of function shifts the membrane balance between endosomes and lysosomes , resulting in the facilitated degradation of endocytosed rhodopsin . Our results demonstrate that the existence of negative regulation in vesicular traffic between endosomes and lysosomes . This mechanism may trigger retinal degeneration in phototransduction mutants .
Drosophila norpA encodes eye-specific phospholipase C and acts as a central effector in phototransduction [6] . The norpA photoreceptor has been used as a model system for studying progressive retinal dystrophies in humans because the loss of its function leads to rapid light-dependent retinal degeneration as a result of excessive endocytosis of stable rhodopsin-arrestin complexes and accumulation of internalized rhodopsin in late endosomes [11] , [12] , [14] . Previous studies [29] have shown that the norpAP24 ( a strong hypomorphic allele of norpA [30] ) photoreceptor showed progressive retinal degeneration ( Figure 1A–B ) . Its degenerative phenotype appeared within three days and was obvious within four days upon constant light exposure . We found that wild-type ( Canton-S ) flies showed no sign of retinal degeneration even after seven days of constant light exposure ( Figure 1F ) , indicating that norpAP24 degeneration was strongly dependent on light . The die4 mutant was previously identified as a norpAP24 suppressor from a genetic screen by using eye-specific FLP-FRT mosaic flies [31] , delaying degeneration several days ( Figure 1G ) . The die4 mutant was generated using ethyl methanesulfonate ( EMS ) mutagenesis , possibly bearing multiple mutations . In addition , mosaic screening enables identification of both lethal and non-lethal mutations . Because of these complexities , we used multiple mapping methods to identify the exact mutation responsible for the suppressive phenotype of the die4 mutant . We previously reported that the mutation in the cytological region of 32D5 to E4 of the die4 chromosome , is responsible for the suppressive phenotype [31] . Although the die4 chromosome is homozygous lethal , this mutation is viable , in that the genomic deficiency , Exel6028 , failed to complement the suppressive phenotype of die4 , and was still viable ( Table S1 ) . In this context , we performed a complementation test of die4 with 11 genes deleted in Exel6028 to identify EMS-induced mutations responsible for the suppressive phenotype ( Figure S1A , Table S2 ) . We identified that the loss of Osi21 ( CG14925 ) is responsible for the suppressive phenotype of die4 , in that the Minos transposon-inserted allele of Osi21 , Mi{ET1}Osi21 [32] , failed to complement die4 and the introduction of the genomic fragment encompassing Osi21 reversed the suppressive effect of die4/Mi{ET1}Osi21 at the deep pseudopupil ( DPP ) level ( Figure 1G , Table S2 ) . Sequence analysis of the die4 chromosome revealed significant amino acid changes ( G149S , M181T , and F229L ) in Osi21 ( Figure 2 , Figure S1 ) . These results were confirmed by targeted knock-down of Osi21 using RNAi method ( Figure 1C ) . Therefore , the suppressive effect of die4 on norpAP24–triggered retinal degeneration is due to the loss of Osi21 function . We therefore conclude that die4 is a loss-of-function allele of Osi21 . Osi21 was identified as an Osiris family protein , without known cellular functions , on the basis of sequence homology [26] . Computational analysis , as described by Shah et al . [27] , was performed using its amino acid sequence , which revealed that OSI21 includes ( 1 ) an endosome/lysosome sorting signal , ( 2 ) a two-Cys region , ( 3 ) duf1676 ( Pfam family: PF07898 ) , and ( 4 ) a YXXØ motif ( Figure 2 ) as predicted by previous studies [26] , [27] , [33] . Interestingly , Osi21 is located on the 2L chromosome . Thus , Osi21 is not linked to the cluster of 20 Osiris family genes that are located in the Triplo-lethal region ( Tpl ) of the 3R chromosome and is responsible for the Tpl phenotype , suggesting that its cellular function differs from that of the other Osiris family proteins . Drosophila rdgC encodes rhodopsin-specific phosphatase and requires rhodopsin recycling [34] , [35] . The loss of rdgC function leads to light- and age-dependent retinal degeneration as a result of excessive endocytosis of stable rhodopsin-arrestin complexes [14] . Because the rdgC mutant photoreceptor cells share the cause of retinal cell death with norpAP24 , we used rdgC306 , the loss-of-function rdgC mutant , to test the effect of die4 on retinal cell death . DPP analysis and histological analysis using electron microscopy showed that die4 protects retinal degeneration due to rdgC306 ( Figure 3A–C , Figure S2 ) , suggesting that Osi21 is not a specific regulator in norpA-triggered retinal degeneration but plays an essential role in retinal degeneration caused by intracellular accumulation of cytotoxic rhodopsin . We also examined the effect of die4 on retinal degeneration due to the loss of trp function , a light-sensitive Ca2+ channel [36] . Drosophila trp1 was recovered as a spontaneously occurring temperature-sensitive loss-of-function mutant at a temperature of 24°C [36] , [37] and is known to show light-enhanced retinal degeneration [38] . Although a dysfunction in Ca2+ fluctuation was suggested as a cause of its retinal degeneration phenotype , its mechanism of degeneration remains unknown . Interestingly , we found that the loss of die4 function protects retinal degeneration caused by trp1 at the DPP and ultrastructural levels ( Figure 3D–F , Figure S2 ) . This result suggests that intracellular accumulation of cytotoxic rhodopsin also causes retinal degeneration in trp1 mutant photoreceptor cells . We used rdgB2 photoreceptors as a negative control to assume the functions of die4 because cytoplasmic rhodopsin aggregation is not involved in retinal degeneration in rdgB2 photoreceptors [39] . As expected , die4 was unable to suppress rdgB2-triggered retinal degeneration ( Figure 3G ) . Combined together , these double mutant analyses suggest that intracellular rhodopsin aggregation triggers unregulated cell death in norpAP24 , rdgC306 , and trp1 photoreceptors , and that Osi21 is a key regulator in the formation of rhodopsin aggregation , wherein the loss of Osi21 function suppresses retinal degeneration in these mutant photoreceptor cells . Previous studies found that norpA and rdgC mutant photoreceptor cells die due to excessive endocytosis of rhodopsin-arrestin complexes and accumulation of endocytosed rhodopsin in late endosomes [11] , [12] , [14] . These findings indicate that the inability of rhodopsin transport and degradation through the endo-lysosomal system cause unregulated cell death in norpA and rdgC mutant photoreceptors . In this context , we tested the possibility that Osi21 acts as a regulator that maintains membrane homeostasis between endosomes and lysosomes in which the functional loss of Osi21 shifts the membrane balance of the endo-lysosomal system . To test our hypothesis , we examined whole mounts of Drosophila retinas by confocal microscopy . We found that the loss of Osi21 function minimally affected the Rab5-positive vesicles ( early endosomes ) ( Figure 4A–B ) and didn't affect the Rab6-positive vesicles ( Golgi complexes ) ( Figure 4C–D ) . However , the loss of Osi21 function significantly affected the Rab7-positive vesicles ( late endosomes ) . Compared to the wild-type photoreceptor cells ( Figure 4E–F ) , both size and number of Rab7-positive vesicles were greatly reduced in Osi21 knock-down photoreceptor cells ( Figure 4G ) . Accordingly , lysosomal compartments proliferated in Osi21 knock-down photoreceptor cells ( Figure 4I–J ) , suggesting that the membrane balance of endo-lysosomal trafficking shifted toward lysosomes . Interestingly , the loss of Osi21 function only affected the number , but not size , of the lysosomal compartments in Osi21 knock-down photoreceptors , thus reflecting limited lysosomal rhodopsin flow in newly eclosed flies . The reduced Rab7-positive vesicles in Osi21 knock-down photoreceptor cells may be attributed to Gal4 titration due to the existence of a second UAS promoter of in the Osi21 knock-down construct . Thus , we used the w; Rhi1::Gal4 , UAS::YFP-Rab7/+; UAS::RFP-arf72A/+ as a titration control and showed that the second UAS promoter did not affect YFP-Rab7 expression ( Figure 4F ) . Quantification of each vesicle clearly showed that among the Rab5- , the Rab6- and the Rab7-positive vesicles , the loss of Osi21 function only affected the Rab7-positive vesicles ( Figure 4K ) . Minimal increase of the Rab5-positive area in Osi21 knock-down photoreceptors may be the secondary effect caused by a reduction in Rab7-positive vesicles . Accordingly , biochemical analysis of isolated endo-lysosomal vesicles using Iodixanol density gradients ( See Text S1 ) showed that immunoreactivities of Rab7 , which was colocalized with endocytosed rhodopsin , were shifted toward the lower density fractions by the loss of Osi21 function ( Figure S3 ) , suggesting a reduced fraction of late endosomes in Rab7-positive vesicles [40] , [41] . These results suggest that the loss of Osi21 function specifically affects the membrane balance between late endosomes and lysosomes . Because the specific shift of membrane balance between late endosomes and lysosomes raised a strong possibility of direct regulation of Osi21 on membrane homeostasis of the endo-lysosomal system , we examined the subcellular localization of the OSI21 protein . Newly eclosed flies were reared in a light/dark cycled incubator and were exposed to bright light ( 2900 lux ) for 90 min to induce massive rhodopsin endocytosis and its accumulation in late endosomes . The subcellular localization of OSI21-GFP from whole mount ommatidia was examined using confocal microscopy . We assumed that the OSI21-GFP is functional because its expression counterbalanced the suppressive effect of Osi21-RNAi in the DPP level ( Data not shown ) . We found the OSI21-GFP localization partially overlapped with Lysotracker staining ( Figure 5A and E , Pearson's correlation coefficient: 0 . 617 ) . In addition , majority of Osi21-GFP colocalized with endocytosed Rh1-RFP ( Figure 5B and E , Pearson's correlation coefficient: 0 . 635 ) , suggesting that Osi21 functions directly on the endo-lysosomal membrane system in a way that Osi21 negatively regulates late endosomal membrane traffic toward lysosomes , resulting in rhodopsin accumulation in late endosomal compartments . Changes in membrane balance between late endosomes and lysosomes may also affect the dynamics of vesicular traffic and the rate of rhodopsin degradation , in which the loss of Osi21 function facilitates rhodopsin traffic toward lysosomes and its lysosomal degradation , resulting in a delay of retinal degeneration in norpAP24 photoreceptor cells . In fact , reduced rhodopsin content due to vitamin A deprivation or mutation in the rhodopsin gene rescued norpA-triggered retinal degeneration [42] . Our analysis by confocal microscopy showed that , compared to the norpAP24 photoreceptor cells ( Figure 5C ) , norpAP24 mutant photoreceptor cells with the Osi21-RNAi transgene showed greatly proliferated lysosomes ( Figure 5D–E ) . These lysosomes were colocalized with endocytosed rhodopsin ( Pearson's correlation coefficient: 0 . 604 ) . Although we often found that small amounts of endocytosed rhodopsin escaped Osi21 blockage ( Figure 5C , arrowhead ) and was colocalized with the lysosome ( Pearson's correlation coefficient: 0 . 342 ) , there were less lysosomes in the control norpAP24 photoreceptor cells . Moreover , majority of lysosomes did not colocalized with endocytosed rhodopsin , indicating that such colocalization reflected regular lysosomal turnover . These results raise the possibility that the loss of Osi21 function facilitates the rhodopsin degradation in lysosomes . In this context , first , we examined the rate of rhodopsin endocytosis and degradation by time course measurements of pulse-chased photoreceptors by confocal microscopy . For the measurements , we expressed RFP-tagged rhodopsin under the control of the hs::Gal4 driver . Newly eclosed norpAP24 and norpAP24; Osi21-RNAi flies were kept in complete darkness for 24 h and then subjected to heat-shock . These flies were then kept in the darkness for another day to allow synthesis and transport of Rh1-RFP to the rhabdomere , following which they were exposed to bright light to chase Rh1-RFP . Whole mount photoreceptors were examined by confocal microscopy at 24 h interval for 96 h by typing the photoreceptor based on the progression of rhodopsin endocytosis and degradation ( Figure 6A ) . We found that the initial movement of endocytosed rhodopsin toward the endosomal system was not different between the norpAP24 and norpAP24; Osi21-RNAi photoreceptor ( Figure 6B , see the percentage of Type I and Type II photoreceptors ) . However , norpAP24 mutant photoreceptor cells with the Osi21-RNAi transgene showed facilitated clearance of endocytosed rhodopsin ( Figure 6B , see the percentage of Type III and Type IV photoreceptors ) , indicating facilitated degradation of endocytosed rhodopsin through the loss of Osi21 function . The effect of the Osi21 loss-of-function on the rhodopsin contents in norpAP24 photoreceptors was also examined by western blot analysis by using flies eclosed within 12 h . Flies were reared either in the dark to prevent rhodopsin endocytosis or in 18 h light/8 h dark cycles to stimulate rhodopsin endocytosis . No significant differences in rhodopsin content were observed in the dark-reared norpAP24 mutant photoreceptor cell with the Osi21-RNAi transgene , compared to dark-reared norpAP24 photoreceptors ( Figure 6C , lanes 2–3 ) . However , the rhodopsin content was greatly reduced in the norpAP24 mutant photoreceptor cells with the Osi21-RNAi transgene by the bright light stimulation ( Figure 6C , lanes 4–5 ) . These results suggest that more endocytosed rhodopsin was transported into and degraded by lysosomes because of Osi21 loss-of-function . Massive influx of Rh1 into the endosomal system may saturate endosomal trafficking machinery , resulting in the late endosomal accumulation of endocytosed rhodopsin . Therefore , we tested the effect of the Osi21 loss-of-function on rhodopsin content when the endosomal trafficking machinery was activated by overexpressing Rab5 or Rab7 ( Figure S4 ) . To induce the maximal rhodopsin endocytosis , newly eclosed flies were exposed to bright light ( 2900 lux ) for 48 h . No significant influence of Rab5 overexpression on the degradation of endocytosed rhodopsin was observed ( Figure 6D , left ) . However , overexpression of Rab7 synergistically accelerated rhodopsin degradation with the loss of function of Osi21 ( Figure 6D , right ) . Considering no significant decrease in rhodopsin content was observed with Rab7 overexpression alone , our results suggest that Osi21 negatively regulated rhodopsin transport between late endosomes and lysosomes by interacting with the Rab7-positive trafficking machinery . Therefore , we conclude that Osi21 is a critical negative regulator of vesicular traffic between endosomes and lysosomes . Its functional loss suppresses retinal degeneration in phototransduction mutants by changing the membrane dynamics between late endosomes and lysosomes and by facilitating the degradation of endocytosed rhodopsin .
In both vertebrates and invertebrates , malfunctioning of phototransduction may stimulate the cell death machinery , resulting in retinal degeneration [43] . Without functional phototransduction , rhodopsin-1 , the major visual pigment , is rapidly endocytosed and accumulated in the late endosomes [12] . Impaired lysosomal delivery of endocytosed rhodopsin and its degradation trigger progressive and light-dependent retinal degeneration in phototransduction mutants [12] , [17] . However , the mechanism underlying the accumulation of endocytosed rhodopsin in late endosomes , instead of delivering to lysosomes for degradation , remains to be elucidated . In the current study , we used die4 , the norpAP24 suppressor , to investigate the molecular basis of the accumulation of rhodopsin in late endosomes in phototransduction mutants . We found that the loss of die4 function delays retinal degeneration in norpAP24 , rdgC306 and trp1 , but not in rdgB2 . These results suggest that , at least , norpAP24 , rdgC306 , and trp1 photoreceptor cells die through a shared route . Previous research suggested that the generation of stable rhodopsin-arrestin complexes is the major cause of cell death in norpAEE5 [11] and rdgC306 [14] . The formation of stable rhodopsin-arrestin complexes in the norpA mutant photoreceptor is attributable to its inability to activate the calcium-dependent phosphatase , RDGC , which dephosphorylates rhodopsin ( Figure 7 ) . The calcium-dependent phosphatase also remains inactive in the trp1 photoreceptor upon light exposure since the cation specific calcium channel is lost in trp1 [36] . Therefore , all three phototransduction mutants share a common feature; the formation of stable rhodopsin-arrestin complexes . On the other hand , norpAP24 , rdgC306 and trp1 require light activation of rhodopsin but not subsequent phototransduction for retinal degeneration [44] . In contrast , rdgB2 requires both , whereby ( 1 ) rdgB2 flies fail to degenerate in complete darkness [44] , ( 2 ) the rdgB2 retinal degeneration is rescued by norpAP24 [44] , and ( 3 ) the rdgBKS222 retinal degeneration is rescued by trp1 [38] . These findings are used to infer that rdgB2 photoreceptor cells die via a different route . We found that the loss of die4 function delays retinal degeneration in norpAP24 longer than those in rdgC306 and trp1 . These results suggest that the blockage of endo-lysosomal trafficking by Osi21 is not the sole cause of retinal degeneration in rdgC306 and trp1 mutants . Recently , Sengupta et al . [45] proposed that PI ( 4 , 5 ) P2 depletion by NORPA underlies retinal degeneration in trpCM and trp343 mutants . Interestingly , both mutants exhibit faster light-dependent retinal degeneration than trp1 mutants . Preventing the formation of stable Rh1-Arr2 complexes by red light slows down the retinal degeneration in trpCM and trp343 mutants comparable to the trp1 degeneration in white light , suggesting that the endocytosis of Rh1-Arr2 complexes contributes retinal degeneration in trp mutants and different results are attributable in part to the allelic differences . In addition , PI ( 4 , 5 ) P2 depletion affects arrestin-mediated endocytosis [45] , so that Rh1 internalization might be reduced in trp mutants . However , their results raise a strong possibility that prolonged activation of NORPA possibly contributes to degenerative syndromes in both trp and rdgC mutants . Double mutant photoreceptor cells are eventually degenerated; they lost their DPP with extended exposure to bright light . Although DPP analysis does not provide a measure of the retinal degeneration process , it faithfully measures a complete loss of the ommatidial structure as it reaches the end of the degenerative process . DPP analysis in the current study suggests that the loss of Osi21 function delays the onset of retinal degeneration in norpAP24 , rdgC306 and trp1 mutants . However , the loss of Osi21 function slows down the retinal degeneration in norpAP24 , but not in rdgC306 and trp1: the slope of DPP loss was similar to that of the control soon after DPP loss occurred in double mutants . These results also suggest that the activity of Osi21 is not the sole cause of the rdgC and the trp degeneration . We conclude that Osi21 acts as a negative regulator of endo-lysosomal membrane traffic between late endosomes and lysosomes . This conclusion is based on the following observations: ( 1 ) Both the size and number of the late endosomes are significantly reduced in Osi21 knock-down photoreceptor cells , ( 2 ) the lysosomal compartments are greatly proliferated in Osi21 knock-down photoreceptor cells , ( 3 ) the OSI21 protein is localized in the endo-lysosomal compartments , ( 4 ) the loss of Osi21 function in the norpAP24 mutant photoreceptor facilitates the degradation of endocytosed rhodopsin , and ( 5 ) Rab7 overexpression alone fails to affect the rhodopsin content of the norpAP24 photoreceptor . However , overexpression of Rab7 synergistically accelerates rhodopsin degradation with the loss of Osi21 function , suggesting that Osi21 directly interacts with the Rab7-positive trafficking machinery . These results clearly demonstrate that the existence of negative blockage regulating the membrane balance of the endo-lysosomal system , and not the capacity of endo-lysosomal trafficking machinery , causes retinal degeneration in phototransduction mutants . Heptahelical G protein-coupled receptors ( GPCRs ) are considered the most diverse and therapeutically important family of receptors [46] , [47] . Like many vertebrate GPCRs , light-activated rhodopsin-1 in Drosophila is rapidly phosphorylated by a specific kinase , called rhodopsin kinase . Phosphorylated rhodopsin-1 is desensitized by arrestins and is endocytosed to terminate further signaling activity ( Figure 7 ) . Because of this similarity , Drosophila rhodopsin-1 has been used as a prototype to study agonist-induced desensitization and internalization of vertebrate GPCRs [48] . In Drosophila photoreceptors , Arr2 promotes rhodopsin endocytosis and degradation when stable Rh1-Arr2 complexes are generated by loss of norpA or rdgC while Arr1 promotes rhodopsin endocytosis and recycling in the normal condition [14] , [49] . Although Arr1 was previously reported to localize in endosomes [49] , we found that Arr2 was absent in the endosomal system ( data not shown ) , indicating Arr2 dissociates from Rh1 near the rhabdomeric membrane . This is reminiscent of functional classification of vertebrate GPCRs: Class A and Class B [50]–[52] . Thus , it can be clearly surmised that Drosophila photoreceptors operate two separate mechanisms of Rh1 endocytosis: ( 1 ) Arr1 for quenching and recycling , and ( 2 ) Arr2 for quenching and degradation . Since Arr2 is several folds more abundant than Arr1 in Drosophila photoreceptor cells to ensure rapid quenching of rhodopsin signaling for visual sensitivity [53] , [54] , the negative blockage by Osi21 may be evolved to counterbalance excessive Rh1 degradation as a result of Arr2 binding to activated Rh1 . Therefore , it should be further investigated whether arrestins play roles in the decision between recycling and degradation for endosomal Rh1 , and in the activation of cell death machinery . Post-endocytic trafficking of GPCRs implicates in many human diseases [55] . Especially , stable rhodopsin-arrestin complexes in vertebrates also result in photoreceptor degeneration [56] , [57] . In addition , cytoplasmic accumulation of proteins often implicates various neurodegenerative disorders , including the accumulation of rhodopsin in retinitis pigmentosa [56] and the accumulation of polyQ-expanded huntingtin in Huntington's disease [58] . Our results suggest that Osi21 regulation may underlie accumulation of disease-causing proteins in the endosomal compartment and that the elimination of Osi21 regulation may clean up this pathologic accumulation . Therefore , the identification and characterization of this specific cellular machinery may provide a therapeutic target for many GPCR-related human diseases and neurodegenerative disorders .
Drosophila was grown on standard food in a 25°C incubator . Standard genetic schemes were used to generate flies with the genotypes described . The Canton-S and w1118 fly were used as a wild-type strain , norpAP24 , rdgC306 , trp1 and rdgB2 as loss-of-function strains of phototransduction . Genomic deficiencies listed in Table S1 . Loss-of-function mutants used for complementation test listed in Table S2 . A w norpAp24 eyFLP chromosome was made using meiotic recombination to subsequently generate die4 mosaic flies in the norpAp24 background . P[ry+; w+]30C , P[ry+; hs-neo; FRT]40A , and P[w +]70C FLP recombinase target ( FRT ) , die4 chromosomes and FRT40A GMR-hid were used in combination with the w norpAp24 eyFLP chromosome to make photoreceptor cells exclusively homozygous for the die4 FRT chromosome [59] . The second and third chromosome inserts of the Rh1::GAL4 driver were derived from the Rh1::GAL4 line constructed by Tabuchi et al . [60] and used for driving expression of various UAS targets including the die4 knock-down construct , UAS::Osi21-RNAi and fluorescently subcellular markers , UAS::YFP-Rab5 , UAS::YFP-Rab7 and UAS::GFP-Rab6 . UAS::Rh1-GFP was constructed using the prh1::eGFP construct from Pichaud and Desplan [61] . All Drosophila stocks except Osi21 knockdown strain were obtained from the Bloomington Stock Center at Indiana University . The UAS::Osi21-RNAi strain was obtained from Vienna Drosophila RNAi Center ( VDRC , Vienna ) . Primers for Osi21 or ninaE ( Rh1 ) were specifically designed for use in the Gateway system ( Invitrogen , Inc . , Carlsbad , CA ) . Exact primer sequences for the rescue experiment , the expression of GFP-tagged Osi21 or RFP-tagged Rh1 were listed in Table S3 . Directional cloning into the pENTR TOPO vector and the destination vectors ( Carnegie Institution of Washington ) followed manufacturer's instruction . The pTW , pTWG , and pTWRvector were used as destination vectors for Rescue constructs , UAS::Osi21-GFP , and UAS::Rh1-RFP , respectively . Plasmid isolation was performed from positive clones using the Qiagen Midi Kit ( Valencia , CA ) . After injection , G0 flies were crossed with w; SM1/Sco; TM2/Sb balancer flies . The progeny from the cross were sorted for mini-w+ eyes . Mini-w+ was used as a marker to determine the presence of the transgene . Flies with the mini-w+ eye ( G1 generation ) were subsequently crossed with w1118 flies to map the location of the transgene . After mating with w1118 flies , mini-w+ flies were crossed to the driver stock . In all cases , mini-w+ flies were crossed to w; Rh1::GAL4 to drive expression of the transgene . The deep pseudopupil ( DPP ) was visualized in red-eyed flies which shows a bright trapezoidal structure when a white light illuminates the retina from the back of the head [62] . The flies of each genotype were collected daily and raised under the appropriate light condition . The flies were also scored daily for the presence of the deep pseudopupil . During degeneration , deep pseudopupils become increasingly diffused before being completely lost . The deep pseudopupil was scored as negative as soon as its trapezoidal shape became indistinct . The percentage of flies that retained their deep pseudopupils for a given day was calculated . In figure 1 , total 20 flies were analyzed for each genotype tested . In figure 2 , three replicates with a total of 100 flies were analyzed for rdgC306 , trp1 and rdgB2 with or without die4 to determine the average percentage of deep pseudopupil -positive flies and the standard error for each day . Detailed procedures were also described previously [63] . Flies eclosed within 6 h were sacrificed with or without light treatment . For the whole-mount ommatidia isolation , fly heads were removed from bodies . A sagittal cut was made on the fly head creating two halves . The brain and proboscis were then removed . The eye was placed on a microscope slide containing 1× PBS . For Lysotracker ( Invitrogen , Inc . , Carlsbad , CA ) staining , fly eyes were preincubated in ∼1 µM Lysotracker for 90 min and washed 3 times in 1× PBS for 30 min . Then , 2% paraformaldehyde in 1× PBS was used as a fixative and treated for 30 min followed by twice wash in 1× PBS . Residual pigments in fly retina were eliminated in 0 . 1% Triton X-100 ( Sigma , MO ) for 4 h , then washed twice for 10 min . Each ommatidia were removed using a sharp platinum needle from fly retina . Usually , a large piece of retina was then teased apart and mounted using mounting medium ( Vector Labratories , Inc . , Burlingame , CA ) . The FV500 confocal laser scanning microscope ( Olympus Optical , Japan ) was used for examining individual ommatidium . Optical images were acquired with an ×100 objective . Confocal images were analyzed with ImageJ software ( NIH , MD ) for quantitative analysis . For quantitative measurement of endosome/lysosome , mean size , number and total area of each vesicle in a photoreceptor cell were calculated with the Analyze Particle function . Measured values were normalized with the known distance option of imageJ . Statistical significances were calculated with two-tailed t tests using Prism 5 . 01 software . For colocalization analysis , Pearson's correlation coefficient ( Rr ) was calculated with Intensity colocalization analysis function of imageJ . The values for Rr range from 1 ( perfect correlation ) to −1 ( perfect exclusion ) . Thus , a value close to 1 indicates reliable colocalization . Retinal degeneration was examined with electron microscopy using retinal tissue sections . Fly eyes were prepared for electron microscopy using procedures described by Washburrn and O'Tousa ( 1992 ) . Electron microscopy sections were ∼80–100 nm thick , stained first in 5% uranyl acetate in 50% EtOH and then in Reynold's lead citrate . The Hitachi H600 electron microscope was used to take electron micrographs . The micrographs shown in all figures are taken from ommatidia cross-sectioned at a depth of R1-R6 photoreceptor nuclei to present a similar view of various genotypes . Usually , 2–5 fly heads were homogenized in buffer A ( 20 mM Tris-HCl ( pH 7 . 5 ) , 100 mM NaCl , 5 mM MgCl2 , 10% sucrose , 1% glycerol , 1 mM EDTA , 1% CHAPS and Complete Protease Inhibitor Cocktail ) with pellet pestle . The homogenate was centrifuged at 4°C and 14 , 000 rpm for 2 min . The supernatant was separated on 12% SDS polyacrylamide gel and then transferred to PVDF membrane at 100 V for 1 hour . The membrane was blocked by 5% nonfat milk in TBST ( TBS with 0 . 5% Tween 20 ) for 60 min . After blocking , the membrane was incubated with the mouse anti-rhodopsin antibody 4C5 ( diluted 1∶5000 , Developmental Studies Hybridoma Bank ) for 60 min at room temperature or overnight at 4°C . The polyclonal rabbit anti-arrestin2 antibody ( Genscript , NY ) is diluted 1∶30 , 000 . The anti-mouse or rabbit IgG HRP conjugated antibody was diluted 1∶5000 in TBST containing 5% skin milk and the membrane was washed with TBST for 30 min . The blotted membrane was detected with a homemade ECL solution for 1 min , and then exposed to X-ray film . The detected bands were quantified using the Quantity One ( Bio-Rad ) . Newly eclosed norpAP24 and norpAP24; Osi21-RNAi flies with UAS::Rh1-RFP under control of hs::Gal4 were kept in complete darkness for 24 h and then subjected to heat-shock for one hour in the 37°C incubator three times at six-hour interval . These flies were kept in the 25°C incubator for another day and then moved under 2900 lux light . Whole mount photoreceptors were examined by confocal microscopy at 24 h intervals for 96 h . At each time point , approximately 30 photoreceptor cells from at least five individuals were scored and categorized as follows: Type I ( most Rh1-RFP localizes in the rhabdomere ) , Type II ( Rh1-RFP localizes equally in the rhabdomere and the cytoplasm ) , Type III ( most Rh1-RFP localizes in the cytoplasm ) and Type IV ( most Rh1-RFP disappears due to degradation ) . The Kolmogorov-Smirnov test for equality distribution was performed using the STATA software package . | Malfunctioning of phototransduction is the major cause of human blindness . Without functional phototransduction , rhodopsin-1 , the major visual pigment , is rapidly endocytosed and accumulated in late endosomes . Impaired lysosomal delivery of endocytosed rhodopsin and its degradation has been reported to trigger progressive and light-dependent retinal degeneration in Drosophila models . It is intriguing why endocytosed rhodopsin accumulates in late endosomes instead of being delivered to lysosomes for degradation . Is this attributable to a saturation of rhodopsin endocytosis , which impedes the delivery capacity of the cell ? To investigate the underlying mechanisms of rhodopsin accumulation in late endosomes , we used a suppressor of phototransduction mutants , which was identified previously from our unbiased genetic screen . This suppressor , called diehard4 , shifts the membrane balance between late endosomes and lysosomes , resulting in the facilitated degradation of endocytosed rhodopsin . Our results clearly demonstrate that a previously unknown mechanism of negative regulation is actively engaged in vesicular traffic between endosomes and lysosomes in fly photoreceptors . We showed that eliminating such blockage alone was enough to rescue retinal degeneration in phototransduction mutants . From these results , we anticipate that the identification of additional components and an in-depth description of this molecular machinery will aid in therapeutic interventions of various retinal dystrophies and neurodegenerative disorders . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"death",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"cellular",
"stress",
"responses",
"animal",
"genetics",
"genetic",
"mutation",
"neuroscience",
"g-protein",
"signaling",
"gene",
"function",
"animal",
"models",
"drosophila",
"melanogaster",
"mode... | 2013 | Negative Regulation of the Novel norpAP24 Suppressor, diehard4, in the Endo-lysosomal Trafficking Underlies Photoreceptor Cell Degeneration |
Human genetics and immune responses are considered to critically influence the outcome of malaria infections including life-threatening syndromes caused by Plasmodium falciparum . An important role in immune regulation is assigned to the apoptosis-signaling cell surface receptor CD95 ( Fas , APO-1 ) , encoded by the gene FAS . Here , a candidate-gene association study including variant discovery at the FAS gene locus was carried out in a case-control group comprising 1 , 195 pediatric cases of severe falciparum malaria and 769 unaffected controls from a region highly endemic for malaria in Ghana , West Africa . We found the A allele of c . −436C>A ( rs9658676 ) located in the promoter region of FAS to be significantly associated with protection from severe childhood malaria ( odds ratio 0 . 71 , 95% confidence interval 0 . 58–0 . 88 , pempirical = 0 . 02 ) and confirmed this finding in a replication group of 1 , 412 additional severe malaria cases and 2 , 659 community controls from the same geographic area . The combined analysis resulted in an odds ratio of 0 . 71 ( 95% confidence interval 0 . 62–0 . 80 , p = 1 . 8×10−7 , n = 6035 ) . The association applied to c . −436AA homozygotes ( odds ratio 0 . 47 , 95% confidence interval 0 . 36–0 . 60 ) and to a lesser extent to c . −436AC heterozygotes ( odds ratio 0 . 73 , 95% confidence interval 0 . 63–0 . 84 ) , and also to all phenotypic subgroups studied , including severe malaria anemia , cerebral malaria , and other malaria complications . Quantitative FACS analyses assessing CD95 surface expression of peripheral blood mononuclear cells of naïve donors showed a significantly higher proportion of CD69+CD95+ cells among persons homozygous for the protective A allele compared to AC heterozygotes and CC homozygotes , indicating a functional role of the associated CD95 variant , possibly in supporting lymphocyte apoptosis .
Severe malaria caused by infection with the protozoan parasite Plasmodium falciparum worldwide causes approximately one million fatalities annually , mostly among children in Sub-Saharan Africa [1] . The clinical picture of severe malaria is characterized by a range of distinct but overlapping syndromes including severe anemia , coma and convulsions , respiratory distress , and others [2] . The variability of the phenotype may be explained by differences in transmission dynamics and the development of the host's immune reactions but also by heritable differences in susceptibility to the disease [3] . The gene FAS ( TNFRSF6 , APT1 ) at chromosome 10q24 . 1 encodes the cell surface receptor CD95 ( Fas , APO-1 ) , known as the prototypic death receptor [4] , [5] . CD95 is a widely expressed molecule with the ability to transduce signals that promote cell death by apoptosis [6] . The CD95-mediated proapoptotic function is triggered by its natural ligand , CD95L , which predominantly is expressed on cells of the T-cell lineage but also acts in a functional soluble form [7] , [8] . The CD95/CD95L system plays a key role in T-cell apoptosis and immune homeostasis as indicated by the induction of lymphoproliferation and autoimmunity in patients with mutations in either the receptor or its ligand [9] . In the case of infections with persistent antigenic challenge , programmed cell death via CD95/CD95L signaling is involved in the elimination of activated lymphocytes , a process indispensable to prevent vital tissues from collateral damage caused by prolonged immune activation [10] , [11] . Indications for an implication of CD95 during severe malaria episodes have been gained from several studies . For instance , lymphocytes from malaria patients were found to express markers of apoptosis and are susceptible to activation-induced cell death ( AICD ) in vitro , and serum samples of patients with P . falciparum malaria show elevated levels of soluble CD95L as compared to healthy subjects [12] , [13] . More specifically , as demonstrated by Balde and colleagues , the exposure of peripheral blood mononuclear cells ( PBMC ) to P . falciparum extract caused a marked increase in the expression of functional CD95 [14] . To date , two studies addressing genome-wide transcriptional changes in blood cells from patients with symptomatic malaria have been conducted [15] , [16] . In Cameroon , the expression profile in PBMC fractions was assessed using samples from adults diagnosed for severe malaria , whereas in a study from Kenya host gene expression was determined in cells from whole blood derived from acute pediatric cases . Although major differences exist between the two experimental designs , both studies detected a significant increase in the expression of CD95 in circulating blood cells during an acute falciparum malaria episode . These findings suggest a role for CD95 in the immune response to infections with P . falciparum , in which its precise function has yet to be defined . In the present study , we sought to elucidate the impact of FAS genetic variants on malaria susceptibility by conducting a candidate-gene association study , which involved a variant screen through re-sequencing and genotyping of selected variants in the FAS gene in a sample set including 1195 severe malaria cases and 769 apparently healthy controls recruited in Ghana , West Africa . In addition , the impact of variant c . −436C>A was further investigated in a replication study from the same geographical area , including 1412 children with severe malaria and 2659 community controls . With regard to the association of variant c . −436C>A ( rs9658676 ) with protection from severe malaria in our study we characterized the allele-dependent CD95 surface expression of PBMCs by quantitative fluorescence-activated cell sorting ( FACS ) .
At total of 19 variants were analyzed in the initial case-control study . Genotypes for 14 of these polymorphisms were derived using the Affymetrix Genome-Wide Human SNP Array 6 . 0 . Additional five variants were selected for genotyping after they were identified by re-sequencing the FAS gene including exonic and regulatory regions in 46 individuals from our study group . Among the 16 single nucleotide polymorphisms ( SNPs ) detected by re-sequencing , two polymorphisms , c . −671G>A ( rs1800682 ) in the 5′-flanking region and c . 46G>A ( rs3218619 ) in exon 1 had previously been shown to exhibit a substantial effect on the expression and function of the receptor [17] , [18] . Hence , both were selected for genotyping aside from c . 365C>T ( rs3218614 ) because it results in a non-synonymous amino acid exchange ( T122I ) in the receptor as well as c . −436C>A ( rs9658676 ) and c . 141G>A ( rs3218621 ) because they showed differences in estimated MAFs between cases and controls of 0 . 16 and 0 . 15 , respectively . One SNP , c . *978C>T , located in the 3′-UTR of the gene , was newly identified . Due to an estimated MAF of 0 . 02 this SNP was not selected for genotyping . The SNPs found , their localizations , and estimated allele frequencies are summarized in Table S1 . Genotype frequencies did not deviate from Hardy-Weinberg Equilibrium ( HWE ) ( p>0 . 01 ) except for c . 334+46C>T , which in the case group showed a deviation with p = 6 . 2×10−4 . When analysing genotypes in the first case-control sample ( n = 1964 ) a significant result was obtained for the promoter variant c . −436C>A in the trend test ( p = 1 . 3×10−3 ) . After adjustment for multiple testing , the association of c . −436C>A in the logistic regression analysis remained significant for the additive model of inheritance ( pempirical = 0 . 02 ) and reached borderline significance for the dominant model ( pempirical = 0 . 05 ) ( Table 1 ) . The A allele of c . −436C>A was found to be more frequent among controls than among cases and thus was associated with protection . The odds ratios ( ORs ) of c . −436C>A were homogeneous in the three ethnic groups included in the study group ( p = 0 . 92 ) . The association of c . −436C>A was confirmed in a replication group of 1412 additional severe-malaria children and 2659 controls ( Figure 1 ) . Like in the initial study group , strongest evidence for association was obtained when applying the additive inheritance model ( OR 0 . 71 , 95% confidence interval ( CI ) 0 . 60–0 . 83 , p = 3 . 1×10−5 ) . Combining the results of the two case-control groups yielded a significance level of p = 1 . 8×10−7 ( fixed effect model; OR 0 . 71 , 95% CI 0 . 62–0 . 80 , Figure 1 ) , with no evidence for heterogeneity of the effects ( p = 0 . 98 , Cochran's Q statistic ) . With a combined OR of 0 . 47 ( 95% CI 0 . 36–0 . 60 ) c . −436AA homozygous individuals appear to be protected to a greater extent than heterozygous individuals ( OR 0 . 73 , 95% CI 0 . 63–0 . 84 ) . The data are compatible with an additive inheritance model ( approach of Bagos , p = 0 . 498 ) , whereas both the dominant and recessive genetic models were rejected ( p = 0 . 044 and p = 0 . 0003 , respectively ) . With the aim of specifying the association of marker c . −436C>A with regard to severe malaria pathology , a stratified analysis for clinically distinct subgroups was carried out . After combining the results from both study groups ORs for the additive mode of inheritance were found to be similar for the clinical subgroups of cerebral malaria , severe anemia , and collectively other complications independent of sample size differences ( Figure 1 ) . Combined ORs ranged from 0 . 68 for cerebral malaria cases ( 95% CI 0 . 42–0 . 94 , p = 1 . 6×10−3; ncases = 569 ) to a maximum of 0 . 73 when including cases with complications such as hyperlactatemia , prostration , and hyperparasitemia collectively ( 95% CI 0 . 58–0 . 92 , p = 0 . 02; ncases = 518 ) . The results pointed towards a protective effect of the A allele in all sub-phenotypes of severe malaria studied . When analyzing log-transformed parasite densities in the combined case groups ( n = 2353 ) , no significant differences between the three genotypic groups were detected ( p = 0 . 67 ) . In order to evaluate a possible association of c . −436C>A with mild malaria , a quantitative transmission disequilibrium test ( qTDT ) was performed with the numbers of uncomplicated malaria episodes experienced by 390 Ghanaian siblings over a period of 31 weeks [19] . No significant distortion of transmission was found ( p = 0 . 96 ) . Likewise , no indication for an association with peripheral-blood parasite densities was detected when using the 75th percentile of the siblings' parasite counts ( p = 0 . 55 ) . Given the clinical and parasitological data , the statistical power to reveal an influence of c . −436C>A on these two phenotypes was 45% and 85% , respectively . Of the additional 18 variants studies ( Table 1 ) , only c . 334+46C>T showed a disease association in the initial case-control group , pempirical being 0 . 04 . This was seen only under a recessive model of inheritance . As the significant findings obtained with c . −436C>A were limited to the assumption of an additive inheritance mode , the association of c . 334+46C>T being of borderline significance was not pursued further . The analysis of the genomic structure at the FAS locus displayed remarkably low pairwise linkage disequilibria between markers , particularly in the promoter region . The variant c . −436C>A shares a maximum r2-value of 0 . 20 with c . 46G>A , located 12 . 6 kb apart in exon 2 of FAS ( Figure 2 ) . Towards the 3′-end of the gene correlation between variants increases , especially for non-coding variants . Within the entire region of approximately 35 kb frequencies of inferred haplotypes were compared in a score test adjusted for age , gender , and ethnicity . Score statistics including full haplotypes with frequencies of >5% did not reveal any significant association with disease ( global p-values: additive model p = 0 . 15 , dominant p = 0 . 32 , recessive p = 0 . 45; haplotype-specific results in Table S2 ) . However , evidence for a haplotypic association was observed when analyzing sub-haplotypes under the additive and dominant models , where a haplotype comprising the three alleles c . −436A , c . 30+1249A , and c . 30+2581T was found to be associated with protection from severe malaria ( additive global p = 0 . 018 , dominant global p = 2 . 2×10−4; Figure S1 ) . Quantitative differences in CD95 surface expression in terms of c . −436C>A genotypes were assessed by staining PBMCs for CD95 from donors of the three genotypic groups . Each PBMC fraction was double-labeled in order to quantify CD95 expression on selected cell types , including CD4+ , CD8+ , CD19+ , and CD69+ cells . Expression levels defined by the median fluorescent intensity ( FI ) for CD95+ cells were similar among the three genotypic groups for both , the entire PBMC fractions as well as the differentiated subpopulations ( Figure 3 ) . However , when examining the proportion of cells expressing CD95 on their surfaces , donors homozygous for the A allele showed a significantly higher percentage of CD69+CD95+ cells than those with heterozygous c . −436AC genotypes ( Mann-Whitney test p = 0 . 027 , Hodges-Lehmann 95% CI 2 . 5–45 . 2 ) or homozygous c . −436CC genotypes ( p = 0 . 048 , 95% CI 2 . 3–47 . 0 ) ( Figure 3 ) . No difference was found between the latter two ( p = 0 . 978 , 95% CI −15 . 1–19 . 6 ) .
Collectively , our results show that the variant c . −436C>A in the promoter region of the FAS gene was associated with protection from severe malaria in Ghanaian children . No differences were found as to the major clinical forms of the disease . Statistical analyses resulted in an OR of 0 . 71 for severe malaria collectively , and ranged between 0 . 68 and 0 . 73 for distinct clinical phenotypes comprising severe malaria anemia , cerebral malaria , and other forms of severe malaria , which in our study included hyperlactatemia , prostration , and hyperparasitemia . It is important to note , however , that the data presented solely allow the conclusion that c . −436C>A is a marker for a haplotype containing one or more genetic variants which reduce the risk of acquiring severe malaria . They do not show that c . −436C>A itself exerts this function . c . −436C>A is not predicted to directly affect a transcription factor binding site ( TRANSFAC 7 . 0 , http://www . gene-regulation . com/pub/databases . html ) . We did not obtain any evidence for additional variants which might be the causal ones by re-sequencing of the genomic region and searching for other variants which are linked to c . −436C>A and might have a stronger association with the phenotypes studied . However , re-sequencing was limited to 46 individuals and the linkage analyses were restricted to a sequence segment of 1 kb and to variants with an MAF of greater than 5% . Therefore , further association signals could have remained undiscovered and additional efforts are needed to convincingly identify the causal variant . c . 334+46C>T , an intronic FAS variant which was associated with marginal significance , was not linked to −436C>A and followed a discordant model of inheritance . A first hint as to the functional effect of the associated genetic variant came from a FACS analysis of PBMCs . Studying apparently healthy donors of the same ethnicity as the participants of the genetic study , the FAS product CD95 was found expressed on CD69+ cells of c . −436AA individuals in a substantially higher proportion of cells than in those from individuals carrying c . −436CA or c . −436CC . These data appear not to be in full agreement with the additive gene effect observed in the association study . CD69 is a cell surface glycoprotein that is considered to be the earliest inducible molecule acquired during lymphoid activation . It is involved in lymphocyte proliferation and functions as a signal-transmitting receptor in lymphocytes , natural killer ( NK ) cells , and platelets [20] . With regard to malaria , in vitro experiments have shown that CD69 is universally up-regulated on NK cells in response to live intact P . falciparum-infected erythrocytes [21] . Thus , the FACS data support the notion that individuals homozygous for the A allele have a higher susceptibility to AICD through a CD95/CD95L interaction . In this experiment PBMCs were used which had been isolated from individuals who appeared healthy by the time their blood was taken . Therefore , it is conceivable to assume that heterozygous c . −436CA individuals have a small increase in the baseline CD95+ cell fraction that was not detectable in the experiment but , during malaria episodes , when CD95 expression is up-regulated [15] , [16] , would show an intermediate phenotype consistent with the additive effect found in the genetic association study . Speculating about a possible mechanism underlying the observed association , it is worth mentioning that in our study the c . −436C>A variants were not associated with differences in parasite counts , neither among our patients during a severe malaria episode nor in a longitudinal observation of 390 Ghanaian children monitored by weekly examinations over 31 weeks . In this cohort , c . −436C>A was not associated with differences in the frequency of mild malaria episodes either , although the latter finding has to be interpreted with caution because of a moderate power to detect significant differences . Nevertheless , the mechanism underlying the c . −436C>A association presumably affects the pathogenesis of the severe forms of malaria . It appears most likely to assume that an increased expression of CD95 associated with c . −436AA and possibly c . −436CA facilitates the programmed cell death of lymphocytes upon immune activation . Thereby the protective FAS allele could alleviate immunopathology . As various studies have shown , immunopathology contributes substantially to the pathogenesis of severe malaria episodes [22] , [23] . Several mechanisms have been proposed to contribute to the pathogenesis of malaria anemia including both , the destruction and decreased production of erythrocytes . Host mechanisms involved in the suppression of erythropoiesis may involve an excessive innate immune response with a persistent production of proinflammatory cytokines [24] . It is possible to envisage a role of CD95 in a down-regulation of cytokine-producing cells through apoptosis , which may alleviate an inhibition of erythropoiesis . Further studies are needed to support these hypotheses and to delineate how they apply to the various forms of malaria complications . With allele frequencies of 0 . 12 in the present study population , 0 . 16 in the Yoruba from Nigeria , and 0 . 09 in African Americans , the A allele has exclusively been found in African populations or those with African ancestry and appears to be absent in Asian and European populations . It is possible to envisage a benefit for individuals carrying the A allele in regions endemic for malaria , which therefore might undergo a positive selection process maintaining the allele in a population . Supportive evidence for this evolutionary aspect comes from a genome-wide search for variants subjected to protozoa-driven selective pressure [25] . In that study , FAS was identified to be among the genes with at least one SNP associated with protozoan diversity in 52 human populations distributed worldwide . Although the protective A allele may have an advantageous effect , with a MAF of 0 . 12 it appears to be maintained at a relatively low frequency in a population constantly exposed to P . falciparum infections . It is possible that its beneficial effect during severe malaria is counterbalanced by an adverse effect of the same allele in other infections . For instance , in patients with AIDS , a disease with too much apoptosis , the depletion of CD4+ T helper cells has been shown to be mediated by CD95 [26] , [27] . In that case the presence of the A allele could lead to a higher apoptosis rate of these cells causing a more rapid loss of peripheral CD4+ T helper cells in these patients . In summary , our study provides a rationale for a more detailed functional characterization of the FAS promoter region , in particular concerning the relevance of the polymorphism c . −436C>A in gene regulation with respect to P . falciparum infections . Further analysis of the CD95/CD95L signaling as part of the immune response to P . falciparum may yield further insights into the pathogenesis of life-threatening childhood malaria .
The studies were approved by the Committee for Research , Publications and Ethics of the School of Medical Sciences , Kwame Nkrumah University of Science and Technology , Kumasi , Ghana . Informed consent was obtained from parents or guardians of all participants for both case-control study groups , and from both parents of families after all procedures were explained in the local language . The present study was carried out in an area highly endemic for falciparum malaria in the Ashanti Region of Ghana , West Africa . The first case-control sample set comprised 1195 severe malaria cases and 769 controls and had previously been subjected to genome-wide SNP genotyping utilizing the Affymetrix Genome-Wide Human SNP Array 6 . 0 ( Affymetrix Inc . , Santa Clara , USA ) ( manuscript in preparation ) . In the herein presented candidate-gene study data for SNPs included in the genome-wide chip were retrieved for the FAS gene locus . In addition , five selected variants based on sequencing results ( see below ) were genotyped in the initial study group . The second , independent sample set derived from the same geographical area included 1412 cases and 2659 controls and was used for the replication of significant results for variant c . −436C>A ( rs9658676 ) observed in the first study . See Table 2 for a summarized characterization of the sample sets . As previously described , all severe malaria patients were enrolled at the Komfo Anokye Teaching Hospital , Kumasi , between 2001 and 2005 in parallel with the “Severe Malaria in African Children” study [28] . Briefly , children aged between 6 and 120 months were included in the study if their Giemsa-blood smear was found to be positive for asexual P . falciparum parasites and either of the following clinical findings was diagnosed: ( i ) level of consciousness according to the Blantyre Coma Score ( BCS ) <3; ( ii ) hemoglobin concentration <5 g/dl; ( iii ) lactate concentration >5 mmol/L [29] . Parasite densities were recorded for 200 leukocytes and calculated assuming a leukocyte count of 8 , 000 per µl blood [30] . Controls in the initial case-control study comprised apparently healthy children who were from the same geographic area and were matched to the case group for age . In addition to apparently healthy children , the controls in the replication study included approximately 10% adults from the same area . In order to determine the impact of the variant c . −436C>A on P . falciparum density in mild childhood malaria , genotypes from an additional study with a total of 739 individuals comprising 390 siblings from 147 families were ascertained . DNA samples from parents and children used here were recruited and prepared as part of a genome-wide linkage analysis conducted in the Asante Akim North District , Ashanti region , Ghana [19] . The phenotype definitions for enrolled children were based on detailed weekly assessments over a period of 7 months during the raining season in 2002 . Following WHO recommendations [31] , mild malaria attacks were defined by either assessing fever ( tympanic temperature of >37 . 7°C ) , or reported fever within the previous 4 days and a positive blood smear for asexual forms of P . falciparum . For each individual , the number of mild malaria episodes during the 31 week observation period was counted , whereby multiple episodes within 3 weeks were counted as one , as they were considered as recrudescences . Among the 390 siblings , a total of 504 malaria episodes were counted during the entire follow-up . Assuming a leukocyte count of 8 , 000 per µl blood parasite counts were recorded per 200 leukocytes ( if >10 parasites/200 leukocytes ) or 500 leukocytes ( if ≤10 parasites/200 leukocytes ) . With a median of 32 parasites per µl blood , parasite densities ranged from 0–317 , 360/µl blood . Due the fact that the median point prevalence of malaria parasites was 54% , the 75th percentile of parasite densities was considered a representative value of parasite density for each individual . Variants located in the coding regions of the FAS gene were detected by re-sequencing the 5′-UTR , the exons , including intron/exon boundaries , and the 3′-UTR of 46 genomic DNA samples drawn from study participants . In order to capture common alleles in the population , 23 DNA samples were selected from the control group . In addition , 23 genomic DNA samples of severe malaria cases were re-sequenced to cover information about possibly selected alleles which confer excess risk for severe malaria . With 23 samples , the probability of observing a variant with a MAF≥0 . 05 in a study group is 90% even if HWE does not hold [32] . Genomic DNA samples were amplified by PCR using primers that captured a 1000 bp region before the transcription start , the exonic sequences , including 30 bp of their intronic flanking regions , and 600 bp of the 3′-UTR . Oligonucleotides were designed using the Primer3 web-interface ( http://frodo . wi . mit . edu/primer3/ ) against the reference sequence ( NCBI NT_030059 . 13 , Transcript NM_000043 . 3 ) . Sequences of oligonucleotides and PCR conditions are listed in Table S3 . After purifying the amplicons using Sephadex G-50 ( Millipore GmbH , Schwalbach/Ts . , Germany ) , each PCR fragment was subjected to a sequencing reaction according to the manufactures' instruction using the BigDye Terminator v3 . 1 Cycle Sequencing Kit ( Applied Biosystems , Darmstadt , Germany ) . Electrophoresis was performed on the ABI PRISM 3100 DNA Analyzer . Assembly of generated sequences against the reference sequence and SNP detection was carried out using the SeqScape Software v2 . 5 ( Applied Biosystems , Darmstadt , Germany ) . The presence of singletons was validated by re-sequencing PCR fragments from both ends . Following the re-sequencing procedure detected variants were evaluated and selected for genotyping in the first case-control study ( n = 1964 ) . Selection criteria were an estimated MAF of ≥0 . 05 in at least one of the two groups , cases or controls , in addition to one of the following attributes . ( i ) The polymorphism leads to a non-synonymous amino-acid exchange in the receptor . ( ii ) The polymorphism is known to have functional relevance based on literature entries . ( iii ) The difference in estimated MAF from the re-sequencing between cases and controls is >10% . In agreement with these conditions a set of 5 SNPs , rs1800682 , rs9658676 , rs3218619 , rs3218621 , and rs3218614 were genotyped in the initial sample set of 1195 severe malaria cases and 769 controls . In order to extract DNA for genotyping , blood samples were drawn from all participants . 0 . 5–1 ml blood was collected into citrate , and , before subjected to DNA extraction , the granulocyte fraction obtained from density gradient centrifugation was preserved in 4-M urea . DNA was extracted according to the suppliers' instructions ( Nucleo-Mag 96 Blood; Macherey-Nagel , Düren , Germany ) . Prior to genotyping whole-genome amplification ( WGA ) using 10 ng DNA of each sample was conducted ( GenomiPhi HY DNA Amplification Kit , GE Healthcare , Braunschweig , Germany ) . SNPs were analyzed by allele-specific hybridization in a melting curve analysis based on fluorescence resonance energy transfer ( FRET ) in a LightTyper device ( Roche Diagnostics , Mannheim , Germany ) . Oligonucleotides and PCR conditions can be found in the Table S4 . For individuals part of the genome-wide association ( GWA ) study genotypes were retrieved for SNPs that are included in the Affymetrix SNP chip and are located at the FAS gene locus or surrounding regions of 10 kb adjacent to start and end of the gene . These were then merged with genotypes from the LightTyper platform . Genotypes were called using Birdseed , and standard quality control procedures comprised the exclusion of ( i ) individuals with SNP-call rates below 0 . 96 , ( ii ) related individuals with an IBD>12 . 5% or ( iii ) individuals with heterozygosity rates <26% or >31% . The cluster plots of individual SNPs were verified by visual inspection . SNPs with call rates <96% were excluded . As part of the standard quality control for the five polymorphisms genotyped on the LightTyper platform , individuals with more than 50% missing genotypes were excluded , and SNPs were allowed to have a maximum of 4% missing genotypes . Hence , a genotypic data set comprising a total of 19 SNPs at the FAS gene locus ( Chr . 10: 90 , 744 , 325–90 , 779 , 097; NCBI human genome Build 37 . 1 ) spanning approximately 35 kb was analyzed using primarily PLINK v1 . 07 [33] . For each polymorphism , the exact test was used to calculate HWE statistics [34] . The Cochran-Armitage trend test and logistic regression analyses were performed with additive coding of genotypes for the initial case-control study . Potential confounding factors , age , sex , and ethnicity were used as covariates in the logistic regression framework . In order to account for multiple testing , empirical significance values were ascertained by using the maxT permutation procedure with 10 , 000 permutations for each tested model , specifically for the additive , dominant , and recessive advantage . Empirical p-values ( pempirical ) <0 . 05 were considered significant . After a significant result was obtained , genotypes for variant c . −436C>A were retrieved in the replication set . We calculated results in the replication sample set by logistic regression accordingly and combined the results using the inverse variance weighted fixed effects model . For each SNP a partitioning χ2-test for heterogeneity of ORs based on stratification into ethnic groups was carried out in order to detect possible population stratification . The most plausible genetic model for variant c . −436C>A was selected using the approach of Bagos [35] . Based on a logistic regression model this approach uses a Wald test in order to test the null hypothesis of equality of the logORs associated with each genotype . Linkage disequilibrium in the gene region was assessed using Haploview v4 . 1 for visualization of pairwise r2 values [36] . Association tests of haplotypes were calculated using the Haplostats package v1 . 4 . 4 in R [37] based on genotypes for all 19 SNPs in the GWA group of individuals ( n = 1964 ) . Score tests were carried out on full inferred haplotypes with frequencies >5% for the additive , dominant and recessive mode of inheritance , as well as on sub-haplotypes , using a sliding window of three SNPs . Age , sex , and ethnicity of individuals were used as covariates in the score statistics . For variant c . −436C>A , the family-based qTDTs were done for the number of mild malaria attacks and the 75th percentile of parasite density using the orthogonal model implemented in QTDT v2 . 5 . 1 [38] . The power calculations for the qTDT analysis are based on a MAF of 0 . 12 and a significance level of 0 . 05 [39] . It was estimated that c . −436C>A genotypes accounted for 5% of variance in the number of malaria episodes and 14% of the variance in the 75th percentiles of parasite density . PBMCs of 72 naïve adult African individuals from the Ashanti region in Ghana were investigated for their level of CD95 surface expression in relation to the c . −436C>A genotype . Prior to specific antibody-labeling and FACS of cells genotypes for c . −436C>A for all 72 individuals were gained as described above . In order to isolate the PBMC fraction from peripheral blood of donors a 2 . 5 ml of citrate blood diluted with 2 . 5 ml RPMI was subjected to Ficoll gradient centrifugation at 4°C . The lymphocyte layer was collected and washed twice with RPMI before PBMCs were stored in freezing medium , RPMI/10% fetal calf serum ( FCS ) /10%DMSO/1×Penicillin/Strepto-mycin at −80°C for 12 hours and subsequently in liquid nitrogen for long-term storage . After thawing cell fractions were washed with RPMI medium and re-suspended in 2 ml of pre-chilled blocking solution ( 1×PBS/10% FCS/10% mouse serum ) . PBMCs were stained with biotinylated mAb against CD95 ( anti-APO-1 IgG1 isotype , P . H . Krammer , Heidelberg , Germany ) as well as with mAbs against CD4 , CD19 , CD8 , and CD69 ( BD Pharmingen , Heidelberg , Germany ) for differentiation of lymphocyte subsets . Biotin-conjugated mouse IgG1 antibody ( BD Pharmingen ) was used as an isotype control . FACS Canto II flow cytometer and FACS-Diva software ( BD Biosciences , San Jose , USA ) were used to perform analyzes of the cells . As cell counts varied significantly with the donors , a minimum of 10 , 000 analyzed cells per assay was set for a sample to be included in the analysis . The surface expression level of CD95 was assessed in two different ways , first by directly measuring FI of cells expressing CD95 and second by defining the percentage of CD95+ cells in particular cell fractions . The average percentage of CD95+ cells and the median FI of donors were compared among three genotypic groups of donors ( c . −436CC , n = 24; AC , n = 22; and AA , n = 26 ) using the Mann-Whitney test at the nominal 5% test-level together with corresponding Hodges-Lehmann 95% confidence intervals . | Severe malaria caused by infection with the protozoan parasite Plasmodium falciparum is a major health burden , causing approximately one million fatalities annually , predominantly among young children in Sub-Saharan Africa . The occurrence of severe malaria may depend on a complex interplay of transmission dynamics and the development of a protective immune response but also on heritable differences in the susceptibility to the disease . In two large studies including a total of 2 , 607 affected children and 3 , 428 apparently healthy individuals from Ghana , West Africa , we investigated genetic variants of the FAS gene , which encodes CD95 , a molecule critically involved in the programmed cell death of lymphocytes . We found that a single nucleotide variant in the FAS promoter was associated with a 29%–reduced risk of developing severe malaria . In individuals carrying two copies of the protective allele , a higher proportion of activated lymphocytes was found to express CD95 . These findings indicate that a predisposition to an increased expression of CD95 may help to protect from severe malaria , possibly by rendering activated T-lymphocytes more susceptible to programmed cell death . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"parastic",
"protozoans",
"plasmodium",
"falciparum",
"genetic",
"association",
"studies",
"genetics",
"protozoology",
"biology",
"human",
"genetics",
"microbiology",
"genetics",
"and",
"genomics"
] | 2011 | A −436C>A Polymorphism in the Human FAS Gene Promoter Associated with Severe Childhood Malaria |
The innate immune response is supposed to play an essential role in the control of amebic liver abscess ( ALA ) , a severe form of invasive amoebiasis due to infection with the protozoan parasite Entamoeba histolytica . In a mouse model for the disease , we previously demonstrated that Jα18-/- mice , lacking invariant natural killer T ( iNKT ) cells , suffer from more severe abscess development . Here we show that the specific activation of iNKT cells using α-galactosylceramide ( α-GalCer ) induces a significant reduction in the sizes of ALA lesions , whereas CD1d−/− mice develop more severe abscesses . We identified a lipopeptidophosphoglycan from E . histolytica membranes ( EhLPPG ) as a possible natural NKT cell ligand and show that the purified phosphoinositol ( PI ) moiety of this molecule induces protective IFN-γ but not IL-4 production in NKT cells . The main component of EhLPPG responsible for NKT cell activation is a diacylated PI , ( 1-O-[ ( 28∶0 ) -lyso-glycero-3-phosphatidyl-]2-O- ( C16:0 ) -Ins ) . IFN-γ production by NKT cells requires the presence of CD1d and simultaneously TLR receptor signalling through MyD88 and secretion of IL-12 . Similar to α-GalCer application , EhLPPG treatment significantly reduces the severity of ALA in ameba-infected mice . Our results suggest that EhLPPG is an amebic molecule that is important for the limitation of ALA development and may explain why the majority of E . histolytica-infected individuals do not develop amebic liver abscess .
Entamoeba histolytica the causative agent of human amebiasis is an intestinal protozoan parasite that causes significant morbidity and mortality worldwide [1] . The main symptoms associated with ameba infection arise when the parasite breach the colonic mucosa , leading to severe hemorrhagic colitis or the development of extraintestinal abscesses , most commonly in the liver . Interestingly , only a small proportion of individuals infected with E . histolytica develop invasive amebiasis while the majority harbors the parasite in the gut without clinical signs of disease [2] , [3] . From in vitro studies as well as from animal models for experimental amebic liver abscess ( ALA ) it is well documented that IFN-γ plays an important role in the early control of E . histolytica invasion . The development of amebicidal activity by neutrophils and monocytes in vitro is dependent on IFN-γ [4]–[7] . Accordingly , the depletion of IFN-γ by monoclonal antibodies or the targeted disruption of the IFN-γ receptor in mice led to more severe tissue destructions of the liver parenchyma in the mouse model for ALA [8] , [9] . Effector lymphocytes with innate like functions such as γδ–T cells , natural killer cells or natural killer T ( NKT ) cells infiltrate into the center of experimental ALA and thus can serve as a source for the protective IFN-γ in the early phase of abscess development . Hence , mice deficient for γδ-T cells , but more importantly , Jα18−/− mice , lacking NKT-cells , have been shown to develop considerably larger abscesses compared to respective controls [8] . NKT cells are involved in immune responses in a broad range of diseases , including autoimmunity , allergy , cancer and infectious diseases [10] . Murine NKT cell populations are heterogeneous [11] , but the majority expresses an invariant Vα14−Jα18 ΤCR and therefore they are referred to as invariant NKT ( iNKT ) cells , that unlike activation of conventional αβ−T cells by antigenic peptides , recognize glycolipids presented by the nonclassical antigen presenting molecule CD1d . Upon ligation of their TCR , iNKT cells can produce large amounts of a variety of cytokines with sometimes opposite function , including the pro-inflammatory IFN-γ as well as the anti-inflammatory IL-4 , IL-10 and IL-13 , which is believed to instruct the development of subsequent immune responses . The prototypical and by far most studied iNKT cell antigen is the glycolipid α-galactosylceramide ( α-GalCer ) , a marine sponge glycolipid , which is a potent CD1d-restricted agonist widely used for in vitro and in vivo experiments to decipher iNKT cell function [12] . In recent years , it has been shown that iNKT cells can be activated directly by recognition of microbial glycolipids presented by CD1d or , indirectly by soluble mediators such as IL-12 and/or by recognition of endogenous ligands , both provided by dendritic cells ( DC ) s−stimulated via Toll-like receptors ( TLRs ) [10] , [13] . In addition to α-GalCer , diverse natural CD1d ligands that stimulate iNKT cells have been identified in various microorgansims [14]–[18] . Accordingly , mice lacking NKT cells and in particular iNKT cells , have an increased susceptibility to various bacterial , fungal , and parasitic infections [8] , [15] , [17] , [19] . Similar to other protozoa , E . histolytica exposes on its surface a complex GPI-anchored glycoconjugate , designated E . histolytica lipopeptidophosphoglycan ( EhLPPG ) [20] . As virulent and non-virulent amebae differ in the amount and the antigenicity of their LPPG these molecules have been associated with pathogenicity [21]–[24] . The structural analysis of EhLPPG revealed the presence of the Gal1Man2GlcN-myo-inositol motif linked to a phosphoserine backbone substituted by the linear carbohydrate chains [Glcα1-6]nGlcβ1-6Gal . Although a lipid anchor was proposed for this molecule , evidence concerning the structure of the lipid has not been provided so far . Thus , information is lacking whether this structure may facilitate a CD1d restricted activation of NKT cells in a CD1d restricted manner . However , toll-like receptor pathways were involved in the induction of IL-12 , IL-8 , IL-10 and TNFα by EhLPPG [25] . The study presented here was aimed to further investigate the role of NKT cells in the development of ALA , to isolate and characterize the structure of a potential natural ligand from E . histolytica trophozoites and to analyse the mechanism of NKT cell activation by dendritic cells presenting EhLPPG .
In a previous study we have shown that mice lacking Vα14-Jα18 iNKT cells have a reduced capacity to control experimentally induced amebic liver abscess ( ALA ) . The lack of control in these knock out mice was evidenced by substantially larger abscess sizes and increased re-isolation rates of E . histolytica trophozoites from the liver lesions when compared to wild type controls [8] . To further investigate the role of iNKT cells in the control of ALA , we specifically activated iNKT cells by using the non-physiological ligand α-GalCer . Wild-type C57BL/6 mice were treated with a single dose of α-GalCer , 24 h prior to ameba challenge . This treatment resulted in a significant reduction of ALA lesions ( p<0 . 003 ) . In contrast , ALA in CD1d−/− mice , lacking iNKT and dNKT cells , were significantly increased , irrespectively whether or not treated with α-GalCer ( p<0 . 006 ) ( Fig . 1 ) . The results indicate the importance of NKT cells in the control of abcess formation . The finding that CD1d is required to limit ALA development suggested that an ameba glycolipid that is presented by CD1d and which is able to activate iNKT cells is involved . A likely candidate is the E . histolytica LPPG ( EhLPPG ) , which is present in considerable quantities on the surface of E . histolytica trophozoites . Accordingly , EhLPPG was isolated using an adaptation of the original reported method [26] . The EhLPPG recovered from the aqueous phase after hot phenol-water extraction was initially characterized by 12% SDS-PAGE gels and was visualized by different staining procedures ( Fig . 2A ) . The purified EhLPPG yielded a negative staining with colloidal Coomassie-blue ( lane 2 ) , which detects as low as 30 ng of protein content . Positive reaction with the periodic acid Schiff ( PAS ) reagent ( lane 3 ) or silver nitrate ( lane 4 ) evidenced a high degree of glycosylation of the molecule resulting in a broad band with two major molecular mass regions between 97–200 kDa and 30–65 kDa , which was in agreement with previous reports [23] , [27] . The presence of EhLPPG was further confirmed by a Western-blot developed with EH5 ( lane 5 ) , a monoclonal antibody specific for EhLPPG [28] . In order to examine whether the purified EhLPPG was able to stimulate lymphocytes from wild-type C57BL/6 mice in vitro , antigen-presenting cells ( APC ) were generated and pulsed with α-GalCer or purified EhLPPG prior to co-cultivation with isolated liver or spleen lymphocytes . Supernatants were analysed for the presence of IFN-γ and IL-4 , respectively . Stimulation of spleen or liver lymphocytes with α-GalCer or EhLPPG resulted in significant IFN-γ production ( Fig . 2B ) . However , similar to recently described microbial glycolipids [17] , the stimulation with EhLPPG reached only 30–50% of the IFN-γ levels induced by α-GalCer , possibly due to the exceptional strong affinity of the non-physiological ligand α-GalCer [29] . In contrast to α-GalCer , EhLPPG-induced lymphocytes did not produce significant amounts of IL-4 ( Fig . 2C ) . The LPS content in the amebic EhLPPG preparations was below 0 . 25 EU/ml at an EhLPPG concentration of 2 µg/ml that induced significant levels of IFN-γ . This LPS concentration did not induce cytokine production in control experiments ( data not shown ) . To further characterize the portion of EhLPPG that might be responsible for lymphocyte activation , and in particular to determine whether the phosphatidylinositol moiety of the GPI anchor from EhLPPG ( EhPI ) is involved , EhPI was separated from EhLPPG after cleavage by nitrous acid deamination . HPTLC analyses demonstrated the presence of two products , designated EhPIa and EhPIb , respectively ( Fig . 3A ) . These bands were isolated and analyzed by GC-MS after methanolysis and peracetylation in order to determine the chemical composition . Both EhPIa and EhPIb contained glycerol ( Gro ) , Inositol ( Ins ) , as well as the fatty acids 30∶1 , 28∶0 and 16∶0 , respectively , the latter being present only in minor amount in EhPIa . ESIFT-ICRMS mass spectrometry revealed the presence of two prominent pseudomolecular ions at m/z 765 . 49 and m/z 739 . 47 in EhPIa , and at m/z 1003 . 72 and m/z 977 . 70 in EhPIb , evidencing a difference of [M]+26 m/z in both cases . The most abundant ions in EhPIa and EhPIb at m/z 739 . 47 and m/z 977 . 70 , respectively , corresponded to the molecules containing 28∶0 , whereas the ions at m/z 765 . 49 ( EhPIa ) and m/z 1003 . 72 ( EhPIb ) corresponded to the molecules containing 30∶1 ( Fig . 3B ) . The fine structural characteristics of both molecules were deduced after ESI-IRMPD experiments ( Fig . 3C ) . For EhPIa , the presence of the daughter ion at m/z 241 . 0 ( inositol-1 , 2-cyclic phosphate ) identified the parent ion as a PI and further indicated that the inositol ring was not acylated at the position 2 [30] . The additional presence of only one fatty acid carboxylate ion at m/z 449 . 4 ( 30∶1 ) and the abundant presence of the ion at m/z 153 . 0 ( cyclic glycerophosphate ) identified this PI as a lyso-acyl anchor [31] . The ions at m/z 585 . 4 and m/z 315 . 0 corresponded to P+Gro+30∶1 and Ins+P+Gro , respectively . In the case of EhPIb , the virtual absence of the ion at m/z 241 . 0 and the presence of the pseudomolecular ions at m/z 585 . 4 ( showing the loss of the Ins+16∶0 fragment ) as well as at m/z 479 . 2 ( P+Ins+16∶0 ) , demonstrated a substitution with 16∶0 at position 2 of the Ins . Acylation of inositol was the only structural feature that differentiated EhPIb from EhPIa . Similar to EhPIa , the abundant presence of the ion at m/z 153 . 0 in EhPIb and the strong fatty acid carboxylate ion at m/z 449 . 4 ( 30∶1 ) , indicated a lyso-acyl anchor in this PI isoform as well [31] . Based on these results , the E . histolytica LPPG lipid anchors were assigned as 1-O- ( 28∶0 ) -lyso-glycero-3-phosphatidylinositol for EhPIa and 1-O-[ ( 28∶0 ) -lyso-glycero-3-phosphatidyl-]2-O- ( 16∶0 ) -inositol for EhPIb ( Fig . 4A ) . Interestingly , stimulation of lymphocytes in the presence of EhLPPG or separated EhPIa and EhPIb , respectively , indicated that EhPIb , but not EhPIa , is the active portion of EhLPPG that contains the capacity to induce the production of IFN-γ ( Fig . 4Β ) . To determine whether indeed NKT cells are responsible for IFN-γ secretion after stimulation with EhLPPG or EhPI , lymphocytes were prepared from Jα18−/− or CD1d−/− mice lacking iNKT or all NKT subpopulations , respectively , and co-cultivated with EhLPPG- or EhPI-pulsed APC from wild-type C57BL/6 mice . Cells from both Jα18−/− and CD1d−/− mice revealed a strong reduction in IFN-γ secretion , indicating that iNKT cells represent the major source for IFN-γ . To further investigate CD1d-restriction of EhLPPG-mediated NKT-cell activation , APC from wild type or CD1d−/− mice were pulsed with α−GalCer , EhLPPG and EhPI , respectively , and co-cultivated with purified T cells from liver or spleen of Vα14 tg mice ( Fig . 5B ) . As expected , in contrast to APC from wild type mice , APC from CD1d−/− mice were impaired in their ability to activate NKT cells . Purified T cells incubated with α-GalCer , EhLPPG and EhPI does not induce IFN-γ in the abscence of APC ( data not shown ) . Taken together , these results suggest that EhLPPG and in particular EhPI activates iNKT cells to produce IFN-γ in a CD1d restricted manner . To determine whether CD1d-restricted iNKT cell activation by EhLPPG requires the involvement of Toll-like-receptor ( TLR ) pathways , APC from knock out mice lacking the TLR-related adapter molecules MyD88 and TRIF , respectively , were pulsed with EhLPPG or EhPI and co-cultivated with T cells from Vα14 tg mice ( Vα14iNKT ) . In addition , parallel experiments were performed with APC from mice lacking TLR1 , TLR2 or TLR6 , or with APC that are impaired to produce functional IL-12 ( IL-12p40−/− ) ( Fig . 6 ) . The results indicated that a TLR pathway other than TLR3 is required for iNKT cell activation , as APC generated from MyD88−/− but not from TRIF−/− mice were unable to induce IFN-γ by Vα14iNKT cells . Moreover , the finding that IFN-γ is induced by APC from TLR1−/− , but not by those generated from TLR2−/− and TLR6−/− mice suggested that EhLPPG or EhPI were capable to activate APC by binding to TLR2 , TLR6 or to TLR2-TLR6 heterodimers . In addition to TLR2 and/or TLR6 signaling , iNKT cell activation by EhLPPG or EhPI required the presence of functional IL-12 , as APC from IL-12p40−/− mice were incapable to induce IFN-γ production by Vα14iNKT cells . We did not detect IL-4 in the supernatants of the co-culture experiments using APC from the various knock-out mutants pulsed with EhLPPG and iNKT cells ( data not shown ) . To gain more insight into the CD1d-restricted activation process we performed several experiments to evaluate if EhLPPG directly binds to CD1d without further processing . To this end EhLPPG was incubated with plate bound CD1d and purified NKT cells . However , no activation of NKT cells was observed . Also CD1d-tetramer incubated with EhLPPG was unable to stain NKT cells ( data not shown ) . These data suggest that EhLPPG cannot bind directly to CD1d molecules . In order to analyse whether EhLPPG enter the endocytic pathway which would allow further processing as a prerequisite for presentation on CD1d , we performed immunofluorescence analysis using the monoclonal anti-EhLPPG antibody EH5 . Here we found that the ameba lipopeptidophosphoglycan is internalized as anti-EH5 staining was found to co-localize with Lamp-1 positive vesicles ( Fig . 7A ) . Thus , EhLPPG was targeted to lysosomes and/or late endosomes which would allow processing and loading of the molecule to CD1d . To corroborate this finding we inhibited the internalization and processing of antigens by pre-incubation of APC with bafilomycin A1 , which interferes with the uptake of macromolecules in endosomes [32] . The inhibition of endocytosis abrogated the IFN-γ production of NKT cells by EhLPPG and α-GalCer , but not the IFN-γ production induced by the TLR-specific activators PamCys or LPS used as positive controls [33] , [34] . In addition the incubation of antigen-pulsed APC with the mAb EH5 , but not with an isotype control ( data not shown ) , inhibited the IFN-γ production of NKT cells in response to EhLPPG ( Fig . 7B ) . This may indicate that the mAb EH5 recognizes EhLPPG presented on CD1d and thus prevented binding to the T cell receptor . Collectively these data indicate that a processing of the EhLPPG molecule is necessary to allow its presentation via CD1d . To investigate whether EhLPPG similar to α-GalCer is able to inhibit ALA development wildtype mice were treated with purified EhLPPG 24 h prior to intrahepatic challenge with virulent E . histolytica trophozoites . The results clearly indicate significant protection in EhLPPG treated mice compared to respective controls ( Fig . 8 ) .
The majority of individuals infected with E . histolytica harbor the parasite within the gut without any clinical signs of disease . Only a small proportion may develop invasive amebaisis , e . g . amebic colitis or amebic liver abscess ( ALA ) . In this study we report on the isolation and characterization of an E . histolytica lipopeptidophosphoglycan ( EhLPPG ) that activates iNKT cells which were previously shown to be crucial in the control of ALA in a mouse model for disease [8] . EhLPPG was found to be a potent stimulator of NKT cells to produce IFN-γ , but not IL-4 . Structural characterization revealed that a particular phosphatidylinositol ( EhPI ) isoform present within the GPI anchor of EhLPPG constitutes the active component for IFN-γ induction . Dissection of the pathway involved in NKT cell activation indicated that direct recognition of EhLPPG or EhPI presented on CD1d was not sufficient for NKT cell activation , as it also required TLR signaling and IL-12 production . The importance of NKT cells for the control of ALA in vivo was previously shown by the use of Jα 18−/− mice lacking iNKT cells [8] and further confirmed in this study by the finding that CD1d−/− mice lacking both iNKT and dNKT cells are more susceptible to ALA development . In addition , application of the most potent and specific NKT cell activator , α-GalCer reduced sizes of ALA significantly . Likewise , mouse models for malaria , trypanosomiasis or cryptococcus infection have shown that a single treatment with α-GalCer rapidly stimulated IFN-γ production by iNKT cells and reduced parasite and bacterial burden , respectively [19] , [35] , [36] . EhLPPG , which is expressed on the surface of E . histolytica trophozoites [20] , [23] , [27] , [37] was identified as the ameba molecule that activates iNKT cells in vitro and which is able to reduce ALA development when applied to mice prior to amebic challenge . To this end we isolated the EhLPPG by an improvement of the original method [26] . This method reduced the possibility of contamination with phospholipids and small glycolipids that could be co-extracted with LPPG and thus may interfere with/or compete for the binding to CD1d , or may potentially provide an indirect stimulus by ligation of TLR . A basic biochemical characterization demonstrated the highly glycosylated nature of the purified compound , which was further identified as EhLPPG by its reactivity with the EhLPPG-specific monoclonal antibody EH5 [28] . EhLPPG is structurally related to glycoconjugates dominating the cell surfaces of other protozoa like Trypanosoma brucei , T . cruzi , Leishmania and Plasmodium . Previous studies on EhLPPG in which the structure was partially characterized indicated the presence of the consensus sequence Gal1Man2GlcN-myo-inositol that corresponds to the core region of GPI anchored molecules , although the nature and arrangement of the lipid portion linked to the inositol moiety remained unclear [27] . In the study presented , it is demonstrated that the EhLPPG is in fact a GPI anchored molecule , applying biochemical degradation , gas chromatography and mass spectrometry of its PI moieties . From these results it is concluded that E . histolytica trophozoites posses two isoforms of PI , EhPIa and EhPIb , respectively ( Fig . 3A ) . In EhPIa , the glycerol is only substituted at the sn-1 position by a single , long fatty acid chain ( 28∶0 or 30∶1 ) , similarly to LPG from Leishmania major [38] . EhPIb bears the same substitution with 28∶0 or 30∶1 at the sn-1 of glycerol but interestingly had an additional 16∶0 in the inositol ring , which is not a common feature among protozoa [39] , although it was found in Plasmodium falciparum GPI [40] and in T . brucei PARP GPI , where the glycerol is monoacylated as well as the inositol is substituted by 16∶0 [41] . This non-common fatty acid distribution might confer particular biological properties to EhLPPG in the interaction with components of the innate immune system like CD1d molecules . These molecules belong to a family of major histocompatibility antigen-like molecules that bind gycophosphinositol with a high affinity and regulate the function and differentiation of NKT cells [42] . In vitro stimulation of APC by EhLPPG induced secretion of IFN-γ by NKT cells . However , compared to α-GalCer , stimulation with EhLPPG was significantly weaker , similar to levels obtained with glycolipid preparations from other microorganisms such as bacteria or protozoa [16]–[18] . The strong induction of IFN-γ by α-GalCer might be due to the composition and length of the alkyl ( C26:0 ) and sphingosine ( C18:0 ) chains , which appear to be optimal for CD1d binding [43] . On the other hand and in contrast to α-GalCer , EhLPPG did not induce secretion of IL-4 . Although the mechanism for the induction of IL-4 is not well understood , a recent investigation suggested that the length of the lipid chain plays an essential role since the truncation of the sphingosine chain promoted the induction of IL-4 and triggered a Th2 immune response [44] . Activation of iNKT cells due to microbial infections can be achieved through direct or indirect pathways ( for rev . see [10] ) . The direct , cognate or antigen mediated activation requires the uptake of the glycolipid , processing and subsequent loading to CD1d molecules in the endosomes of the APC . In this scenario , iNKT cells are engaged via their invariant TCR recognizing antigen presented on CD1d without any contribution of additional cytokines released from the APC . Beside α-GalCer as the classical direct activator of iNKT cells , glycolipids from various bacteria or protozoa have the capacity to activate NKT cells by CD1d restriction independent from cytokines provided by the APC [15]–[18] . For indirect or adjuvant-like activation of iNKT cells , different possible pathways have been proposed . One of the indirect pathways involves the recognition of PAMPS by TLR resulting in the induction of IL-12 and/or IL-18 activating NKT cells in a CD1d-independent manner [45] . During Salmonella typhimurium infection , the NKT cell response depends on IL-12 and additionally on CD1d recognition [46] . Another activation mechanism has been described for LPS which induces IFN-γ production in NKT cells via IL-12 and IL- [34] . A third alternative is an indirect activation pathway induced by parasite eggs . Here endogenous glycolipid is upregulated by a yet unknown mechanism and presented via CD1d to NKT cells . So far , this mechanism has been described only for DCs sensitized with eggs from Schistosoma mansoni [47] . From the results presented here using a series of knock-out mice we conclude that activation of iNKT cells by EhLPPG or EhPI requires both presentation of EhPI by CD1d as well as TLR signalling and IL-12 production . In addition , we provide evidence that endocytosis and processing of EhLPPG is a prerequisite for CD1d-dependent NKT cell activation since i ) we could not find a direct binding of EhLPPG to recombinant CD1d , ii ) EhLPPG was found to co-localize with Lamp-1 as a marker for late endosomes which would allow processing and loading of the molecule to CD1d and iii ) we showed that NKT cell activation is abrogated when APC endocytosis was inhibited by bafilomycin [48] . In addition , the incubation of APC with the mAb EH5 inhibited the IFN-γ production of NKT cells in response to EhLPPG . The finding that induction of IFN-γ by APC from TLR2−/− and TLR6−/− mice is abrogated suggests that EhLPPG and EhPI is capable to activate the APC by binding through these TLR . This is in agreement with recent results that diacylated motifs , as also present in the active EhPIb isoform , bind to TLR2-TLR6 heterodimers , while triacylated molecules engage TLR2-TLR1 heterodimers [49] , [50] . However , we can not exclude that the inability of EhLPPG and recombinant CD1d to stimulate NKT cells is due to a lack of costimulatory molecules and/or cytokines . Moreover , the lack of direct binding of EhLPPG to CD1d might indicate that the observed activation of NKT cells by EhLPPG also involves an increased presentation of self-antigens by CD1d . Interestingly , we did not find activity with the monoacylated EhPIa isoform , although such a structure is in principle capable to activate NKT cells via TLR2 and CD1d dependent pathways as shown for a structurally related LPG from Leishmania [15] , [51] . In consideration of an adjuvant effect of TLR-ligands during activation of CD1d-restricted NKT cell , traces of contaminating lipopeptides could also be responsible for the TLR2-dependent response [52] . However , the double extraction method ( chloroform methanol water and phenol water ) used for the purification of EhLPPG should minimize the risk of contamination [53] . Our data demonstrate that early IFN-γ production by NKT cells is responsible for a sufficient control of parasites in the liver . In addition , NKT cells provide a link between innate and adaptive immunity due to their capacity to early produce large amounts of IFN-γ and IL-4 that can bias the immune response into either a TH1-or TH2 direction . The exclusive IFN-γ production of EhLPPG activated NKT cells can be expected to trigger the subsequent adaptive immune response into a TH1 type that would provide additional IFN-γ . This might augment efficient abcess control by T cell dependent mechanisms at later time points . Taken together the results presented here indicate that EhLPPG is able to limit ALA development most likely due to its ability to specifically activate iNKT cells to produce IFN-γ . The importance of IFN-γ for the control of E . histolytica has been documented in various investigations [4] , [5] , [7] , [8] , [9] , [54] . In particular , the mouse model used in this study recently revealed that the application of IFN-γ - neutralizing antibodies abrogates ALA development . Thus EhLPPG constitutes an ameba molecule that is critically important to control ALA and which might be responsible for the lack of amebic disease in the majority of E . histolytica infected individuals .
Trophozoites of the E . histolytica isolate HM-1:IMSS were grown axenically in TYI-S-3 medium [55] . To maintain virulence , trophozoites were regularly passaged through the liver of C57BL/6 mice as described previously [8] . Wild-type C57BL/6 ( WT ) , Vα14-Jα18 transgenic ( tg ) , TLR1−/− , TLR6−/− , CD1d−/− and Jα18−/− were bred and housed under specific pathogen-free conditions at the Bernhard Nocht Institute for Tropical Medicine ( Hamburg ) . IL12p40−/−; TLR2−/− and MyD88−/− were kindly provided by Christoph Hölscher , Research Center Borstel , Germany . TRIF−/− were kindly provided by Bruce Beutler , La Jolla , California . All mice were backcrossed on a B6 genetic background for >10 generations . Amebic liver abscesses were induced by direct intrahepatic inoculation of virulent E . histolytica trophozoites as previously described [56] with minor modifications for the use of C57BL/6 mice [8] . The influence of α-GalCer and EhLPPG on the abscess formation was investigated by intraperitoneal application of 2 µg α-GalCer ( Alexis , Axxora ) or 4 µg EhLPPG diluted in PBS/0 . 05% Tween 20 per animal 24 h prior to amebic challenge . In our previous work we found that abscesses are self-limited and were resolved until day 21 post infection . Performing a kinetic analysis of abcess formation we found that day seven post infection is the most appropriate time point for studying the influence of immune mechanisms on abcess size [8] . Therefore , on day seven post intrahepatic inoculation of E . histolytica trophozoites , the animals were sacrificed and the size of the abscess lesions were measured in mm , a score was introduced and related to the score of ALA found in WT mice ( score: 0 = no abscess; 1 = <1 mm; 2 = 1–5 mm; 3 = >5 mm ) . Trophozoites of the late logarithmic phase of growth were washed , resuspended in pyrogen free water and lysed by freeze and thawing . The homogenate was centrifuged at 430 g at 4°C for 10 min and subsequently the supernatant was recovered and ultracentrifuged at 150 , 000 g for 40 min [57] . The obtained pellet was extracted with a mixture of chloroform/methanol/water 10∶10∶3 ( by volume ) and the insoluble material was recovered by centrifugation , dried , resuspended in distilled pyrogen free water and extracted three times with an equal volume of 90% phenol at 68°C for 30 min with constant stirring [58] . The water phase containing EhLPPG was recovered after centrifugation at 12 , 000 g for 30 min and dialysis against distilled water . In order to obtain the EhPI moiety , nitrous acid deamination was performed as described [59] . In brief , dried EhLPPG was resuspended in a mixture of 0 . 3 M sodium acetate buffer at pH 4 . 0 and 1 M sodium nitrite , incubated at 37°C for 2 h . The released EhPI moiety was recovered from the organic phase after partition between water and water-saturated 1-butanol . The organic phase was then dried under a stream of nitrogen and resuspended in a mixture of chloroform/methanol/water 10∶10∶3 ( by volume ) for analysis on Silica Gel 60 high-performance thin-layer chromatography ( HPTLC ) plates . EhPIa and EhPIb were separated on a preparative HPTLC and re-extracted with a mixture of Chloroform/methanol/water 10∶10∶3 ( by volume ) . For the use in cell stimulation , PI , PIa and PIb were dried under a stream of nitrogen and resuspened in PBS/Tween ( 0 . 05% ) . EhLPPG and EhPI samples contained <0 . 25 endotoxin units ( EU ) per ml at the concentrations used for cell activation , as determined by the Limulus amoebocyte lysate assays ( Cambrex ) . EhLPPG ( 10 µg/lane ) was analyzed by 12% SDS-PAGE and either stained to evidence the presence of protein ( Coomassie brilliant blue ) and carbohydrates ( silver nitrate and periodic acid Schiff ) , or transferred to a PVDF membrane for Western blot analysis and subsequently developed with the LPPG-specific monoclonal antibody EH5 [28] . For compositional analyses , EhLPPG and EhPI were subjected to methanolysis . 150 µl of 0 . 5 M hydrochloric acid in dry methanol was added and the solution incubated at 85°C for 1 h , followed by addition of 50 µl of both pyridine and acetic anhydride and analyzed by GC-MS on a Hewlett Packard GL 5890 Gas chromatograph equipped with an Ultra-1 column ( Agilent ) and coupled to an electron impact mass detector . Electrospray Ionization Fourier Transform Ion Cyclotron Mass Spectrometry ( ESI FT-ICR MS ) was performed in the negative ion mode using an APEX Qe – Instrument ( Bruker Daltonics , Billerica , USA ) equipped with a 7 Tesla actively shielded magnet . Mass spectra were acquired using standard experimental sequences as provided by the manufacturer . Samples were dissolved at a concentration of ∼10 ng/µl in a 50∶50∶0 . 001 ( v/v/v ) mixture of 2-propanol , water , and triethylamine and sprayed at a flow rate of 2 µl/min . Capillary entrance voltage was set to 3 . 8 kV , and dry gas temperature to 150°C . Infrared-multiphoton dissociation ( IRMPD ) of isolated parent ions was performed with a 35 W , 106 µm CO2 laser ( Synrad , Mukilteo , WA ) . The unfocused laser beam was directed through the center of the ICR cell and fragment ions were detected after a delay of 0 . 5 ms . The duration of laser irradiation was adapted for each sample to generate optimal fragmentation and varied between 10–80 ms . Bone marrow was harvested from femurs of 6- to 10-week-old mice and cultured as described by Lutz et al . [60] . Cultures were supplemented with supernatants from Ag8653 myeloma cells transfected with the gene coding for murine GM-CSF ( kindly provided by B . Stockinger , NMRI , Mill Hill , London , UK ) . The percentage of mature cells was determined by FACS analysis using anti- CD11c - APC , anti - CD40-FITC , anti - CD86-PE or anti - CD80 ( BD-Bioscience ) and ranged from 20–23% . Spleens were perfused with hypotonic ammonium chlorid solution for erythrolysis . Subsequently , cells were washed twice with medium and adjusted to the appropriate number . Livers were perfused with ice cold PBS/20% FCS solution and subsequently filtered through a 40 µm mesh . Following centrifugation at 400 g , cell pellets were resuspended in RPMI-20 medium , underlayed with a 30% Nycodenz solution ( NycoprepTM , Universal ) and centrifuged at 900 g for 20 min . The liver lymphocytes were collected from the interface , treated with hypotonic ammonium chloride solution , washed and subjected to magnetic bead sorting using the Pan T-cell isolation kit ( Macs , Myltenyi ) . NKT-cell populations were analyzed for purity by flow cytometry using α-GalCer ( Alexis; Axora ) loaded CD1d-tetramer-PE ( Proimmune ) and anti-CD4-FITC ( BD Bioscience ) . Purified NKT cells did not produce IFN-γ after cultivation with α-GalCer in the abscence of aditional APC . BMDC were used as antigen presenting cells ( APC ) . In brief , 4×104 cells/well were cultured in triplicates in 96-well round bottom plates with RPMI-20 medium , supplemented with FCS , L-glutamine , antibiotics , sodium pyruvate . The cells were pulsed with α-GalCer or purified amebic EhLPPG or EhPI at indicated concentrations for 3 h . NKT-cell enriched spleen and liver lymphocytes were then added to the pulsed APC with 1×105 cells/well and incubated for 24 h ( IL-4 ) and 48 h ( IFN-γ ) . Cytokines were measured with the respective sandwhich ELISA for IL-4 and IFN-γ ( R&D System ) . To block the processing and presentation pathway , APCs were treated with 10 nM bafilomycin A1 ( Sigma ) 30 min before incubation with EhLPPG , α-GalCer , PamCys-SKK ( EMC , Germany ) or LPS ( E . coli , serotype 055∶B5; SIGMA ) . In an additional set of experiments APC were pulsed for 3 h with EhLPPG or α-GalCer . Subsequently APC were incubated with 100 µg/ml mAb EH5 [28] before the addition of NKT cells . APC from WT BL6 mice were pulsed with 20 µg/ml of EhLPPG and seeded in 24-well plates with glass coverslips and incubated with 5%CO2 at 37°C for 3 h . Subsequently , cells were washed , fixed with 4% paraformaldehyde and permeabilized with 0 . 5% saponin ( Sigma ) . Intracellular EhLPPG was stained with mouse-monoclonal antibody EH5 [28] followed by Alexa 594 labeled anti-mouse IgG ( Invitrogen ) . The internalized EhLPPG colocalized with a rat anti- CD107a/LAMP-1 Mab ( Southern Biotec ) and Alexa 488 labelled anti-rat IgG ( Invitrogen ) . Nuclei were stained with 4‵ , 6‵-diamidino-2-phenylindole hydrochloride ( DAPI ) . All stainings were visualized with a Leica DRM confocal microscope and OpenLab software ( Improvison . Inc ) . Statistics were performed using Prism statistical software ( GraphPad ) unpaired , one-way , non parametric Mann-Whitney U tests , the ANOVA , Dunnett and the student t test . | Amoebiasis is a widespread human parasitic disease caused by the intestinal protozoan Entamoeba histolytica . There are two major clinical manifestations of the disease , amebic colitis and amebic liver abscess . Interestingly , only a small proportion of E . histolytica-infected individuals develop invasive disease , whereas the majority harbors the parasite within the gut without clinical symptoms . So far , cells of the innate immune system have been described to constitute the main host defense mechanism for the control of amoebiasis , relying largely on the early production of interferon-γ ( IFN-γ ) , a protein , which activates macrophages to kill microorganisms including parasites . However , the IFN-γ-producing cells as well as the amebic antigen involved in the activation have not been identified . Here we demonstrate that mice challenged with live E . histolytica , and which are deficient of a specific lymphocyte population known as natural killer T cells ( NKT cells ) , have a reduced capacity to control ameba infection and develop much larger amebic liver abscesses compared to normal mice . In addition , we isolated a molecule from the surface membrane of E . histolytica , which constitutes a lipopeptidophosphoglycan , and which activates NKT cells for the production of protective IFN-γ . Thus , our study provides a mechanism for the innate control of ameba invasion that might explain why the majority of E . histolytica-infected individuals do not develop amebic disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry/macromolecular",
"chemistry",
"immunology/innate",
"immunity",
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/tropical",
"and",
"travel-associated",
"diseases",
"immunology/leukocyte",
"activation"
] | 2009 | Natural Killer T Cells Activated by a Lipopeptidophosphoglycan from Entamoeba histolytica Are Critically Important To Control Amebic Liver Abscess |
Nonsense Mediated Decay ( NMD ) degrades transcripts that contain a premature STOP codon resulting from mistranscription or missplicing . However NMD's surveillance of gene expression varies in efficiency both among and within human genes . Previous work has shown that the intron content of human genes is influenced by missplicing events invisible to NMD . Given the high rate of transcriptional errors in eukaryotes , we hypothesized that natural selection has promoted a dual strategy of “prevention and cure” to alleviate the problem of nonsense transcriptional errors . A prediction of this hypothesis is that NMD's inefficiency should leave a signature of “transcriptional robustness” in human gene sequences that reduces the frequency of nonsense transcriptional errors . For human genes we determined the usage of “fragile” codons , prone to mistranscription into STOP codons , relative to the usage of “robust” codons that do not generate nonsense errors . We observe that single-exon genes have evolved to become robust to mistranscription , because they show a significant tendency to avoid fragile codons relative to robust codons when compared to multi-exon genes . A similar depletion is evident in last exons of multi-exon genes . Histone genes are particularly depleted of fragile codons and thus highly robust to transcriptional errors . Finally , the protein products of single-exon genes show a strong tendency to avoid those amino acids that can only be encoded using fragile codons . Each of these observations can be attributed to NMD deficiency . Thus , in the human genome , wherever the “cure” for nonsense ( i . e . NMD ) is inefficient , there is increased reliance on the strategy of nonsense “prevention” ( i . e . transcriptional robustness ) . This study shows that human genes are exposed to the deleterious influence of transcriptional errors . Moreover , it suggests that gene expression errors are an underestimated phenomenon , in molecular evolution in general and in selection for genomic robustness in particular .
In mammalian transcripts , premature termination codons ( PTCs ) are normally detected by one of two NMD pathways [1] . The primary , and more efficient , NMD pathway is intron-dependent and relies on the presence of exon junction complexes ( EJCs ) deposited 20-24 nts upstream of the exon-exon junctions created following splicing of the mRNA ( EJC-dependent NMD ) . A second , less efficient , intron-independent pathway operates in mammals and requires the presence of polyA-binding protein ( PABP-dependent NMD ) . In contrast , in Drosophila melanogaster , PABP-dependent NMD is the primary pathway in operation since only a minority of spliced transcripts are under the surveillance of EJC-dependent NMD [2] , [3] . The diversity of NMD mechanisms in eukaryotes is further illustrated among yeast species . In Saccharomyces cerevisiae only the intron-independent NMD pathway is known to operate . Finally , in Schizosaccharomyces pombe , although intron-dependent NMD is active it is believed to be EJC-independent [4] . PTCs may arise from heritable nonsense mutations in the germline or they can be created by transient errors in transcription and splicing . Previous attention has been paid to splicing errors as a source of nonsense errors since introns retained in the mature mRNA can either introduce an intron-encoded PTC or induce a frameshift leading to the formation of an exon-derived PTC [5] . However , mistranscription is likely to be a significant source of nonsense errors . Although few direct estimates are available , assuming a transcriptional error rate of 10−5 errors per nucleotide [6] , [7] leads to the estimate that 0 . 05%-0 . 5% of transcripts of any given gene are expected to contain PTCs due to mistranscription [8] . Moreover , this is a highly conservative estimate since the single measurement available for a metazoan suggests that the rate of mistranscription may be as high as 10−3 errors per nucleotide [9] . The possible consequences of translating PTC-containing transcripts include loss of functional molecules , dominant negative interactions and gain-of-function activities . The deleterious effects of these outcomes are highlighted by the fact that , in mammals , knockouts of the core effectors of NMD have lethal effects [10] . NMD mitigates the negative effects of PTC-containing transcripts and hence can be considered as providing a partial “cure” for transcriptional errors . In mammals , the efficiency of NMD in detecting PTCs varies both among and within genes . Single-exon genes appear insensitive to NMD [11] , [12] owing to the fact that these genes are served only by the inefficient PABP-dependent pathway and are invisible to the more potent EJC-dependent NMD pathway . As a result , PTC-bearing transcripts of single-exon genes are more likely to be translated into truncated protein species with toxic effects . Therefore , it follows that transcriptional errors creating nonsense codons are likely to be more deleterious in single-exon than in multi-exon genes . Equally , in multi-exon genes , the EJC-dependent NMD pathway is thought to only detect PTCs lying 50–55 nts upstream of the last exon-exon junction [13] . Consequently , like single-exon genes , last exons of multi-exon genes are invisible to EJC-dependent NMD in the mammalian genome and are served only by the PABP-dependent NMD pathway . In this study we propose that natural selection has promoted a dual–strategy of prevention and cure to deal with nonsense errors in gene expression . Specifically we predict that selection compensates for the partial efficiency of NMD through a preventative approach of “transcriptional robustness” that minimizes the consequences of transcriptional error . We investigated whether human genes eschew codons that can be changed into a stop codon by one mutation ( ‘fragile codons’ [14] ) in favour of codons robust to nonsense errors during mistranscription ( ‘robust codons’ ) . The standard genetic code comprises 18 fragile codons and 43 robust codons ( Figure 1 ) . Of the 20 amino-acids , six are encoded exclusively by fragile codons ( “fragile amino acids” ) , ten are encoded exclusively by robust codons ( “robust amino acids” ) and four can be encoded by codons of either type ( “facultative amino acids” ) . This suggests two distinct mechanisms by which the transcriptional robustness of a gene can be increased by natural selection . First , at the level of synonymous codon usage fragile codons can be avoided when a robust synonym exists . Second , at the protein level fragile amino-acids can be counterselected . Using comparisons between single and multi-exon genes ( intergenic ) and between different exons of multi-exons genes ( intragenic ) , we show how each of these mechanisms is used to increase transcriptional robustness of mammalian genes . Single-exon genes should provide the clearest signal of transcriptional robustness and thus provide a “litmus test” of this hypothesis . Therefore , our first , intergenic , analysis contrasts fragile codon usage and fragile amino-acid usage in single-exon genes and multi-exon genes . We performed a second , intragenic , analysis at the exon-level by taking advantage of the fact that , in mammals , last exons of multi-exon genes , like single-exon genes , are invisible to EJC-dependent NMD . Finally , we demonstrate that the two hallmarks of transcriptional robustness detectable in mammals ( synonymous codon usage and amino-acid usage ) can also be detected in the genes of fission yeast ( Schizosaccharomyces pombe ) in which a mechanistically distinct form of intron-dependent Nonsense Mediated Decay appears to be active .
We used an intergenic comparison to focus initially on the first mechanism to achieving transcriptional robustness and quantified it using a normalized fragile codon usage ( NFCU ) metric . Being mutational neighbours of STOP codons , the 18 fragile codons are relatively AT-rich . It is known that GC-content varies along chromosomes and that these variations in base composition affect both synonymous codon usage and amino-acid usage . These variations are particularly strong within mammalian genomes ( the so-called isochore genome organization ) . To account for both this compositional bias and the influence of amino acid content , we used codon usage among facultative amino acids as the basis of the NFCU metric . NFCU measures the relative usage of fragile and robust codons among 5 groups of codons where all codons in a given group ( i ) have the same GC-content and ( ii ) are synonymous . We measured NFCU in 2422 single-exon genes and in 20563 multi-exon genes in the human genome . We observe a 8% depletion of fragile codons in single-exon genes that is highly significant ( p<10−15 , Wilcoxon rank sum test; Table 1 ) . Similarly , in mouse , measuring NFCU in 3582 single-exon genes and 20263 multi-exon genes reveals a highly significant 11% depletion of fragile codons in single-exon genes . Both results provide preliminary support for our hypothesis . We performed a negative control by repeating the analysis in D . melanogaster in which both gene sets should on average be equally well served by NMD since the intron-dependent NMD pathway shows much reduced activity in fly [2] , [3] . In the fly genome , the depletion of fragile codons in single-exon compared to multi-exon genes is almost null ( 2% ) ( Table 1 ) and , in contrast to human and mouse , can be explained by transcript length differences ( see below ) . Notably , repeating the analysis without controlling for amino acid usage or nucleotide composition but with the benefit of higher coverage of codons , yields essentially the same result ( Text S1 , Result A; Table S1 ) . Interestingly , one group of genes falls entirely outside of the range of NMD's surveillance . Replication-dependent histones contain neither introns in their coding sequences nor polyA-tail in their mRNAs [15] . Therefore , histone genes represent a blind-spot for both mammalian NMD pathways [12] . According to our hypothesis histone genes should represent the most transcriptionally robust genes in the mammalian genome since PTC-containing transcripts of their genes will not be recognized and degraded before translation . In agreement with this , in human we see that NFCU in histone genes is 32% lower than that of multi-exon genes ( Table 1 ) and 26% lower than that of other single-exon genes . Similarly , in mouse we see that NFCU in histone genes is 53% lower than that of multi-exon genes ( Table 1 ) and 47% lower than that of other single-exon genes . Notably , we found that histone genes in fly show a 30% depletion of NFCU compared to other single exon genes in accordance with the fact that histone genes are not served by NMD in fly ( Table 1 ) . The analysis of fragile codon usage controlling for GC and amino-acid usage ( NFCU ) shows that the depletion of fragile codons in human and mouse single-exon genes is not a consequence of amino acid usage . Therefore , we can ask whether there is evidence for usage of the second mechanism of transcriptional robustness by testing whether the amino acid usage of single-exon gene products reduces the frequency of nonsense transcriptional errors . Accordingly , we observed that the usage of fragile amino acids ( FAU ) among proteins encoded by single-exon genes is 17% lower than those of multi-exon genes in human and 21% lower in mouse whereas no difference is observed in fly ( Table 2 ) . Therefore the constraints on codon usage in human genes imposed by the need for transcriptional robustness appear to be strong enough to influence their amino-acid sequences . We repeated this analysis to control for GC content differences between single-exon and multi-exon genes . We used a normalized fragile amino-acid usage metric ( NFAU ) that considers the relative usage of fragile amino-acids among two groups of amino-acids that are encoded by codons having the same GC-content . Once GC content is controlled for we observe that the fragile amino acid content of single-exon gene products is 12% lower than those of multi-exon genes in human and 16% lower in mouse whereas a 2% enrichment is seen in fly ( Table 3 ) . Thus far our analysis has demonstrated that , in human and mouse , single exon genes bear two hallmarks of transcriptional robustness relative to multi-exon genes ( i . e . depletion of fragile codons and amino acids ) . These hallmarks highlight two mechanisms by which natural selection can prevent nonsense transcriptional errors: at the level of gene sequence through synonymous codon choice and at the level of protein sequence through amino-acid choice . Provisionally , we can attribute this observation to the inactivity of EJC-dependent NMD in single-exon genes . However , since all biological processes are inherently inefficient , EJC-dependent NMD is inevitably also suboptimal . This raises the question whether there is a genome-wide requirement for transcriptional robustness . One approach to answering this question is to ask whether , among all human genes , there is correlated usage of the two mechanisms ( codon-level and protein-level ) to achieving transcriptional robustness . However , for a given gene the nature of the correlation between these two mechanisms is likely to depend on the relative level of constraint on synonymous and non-synonymous sites as measured by Ka/Ks ( the ratio of non-synonymous to synonymous substitution rates ) . For the vast majority of human genes , estimates of Ka/Ks lie in the range 0-1 . Broadly , we might expect two different patterns: ( i ) For genes with Ka/Ks close to 1 , synonymous and non-synonymous sites are equally modifiable . Therefore , where nonsense errors are deleterious , natural selection can use both mechanisms to increase transcriptional robustness ( i . e . both fragile codons and fragile amino-acids can be depleted ) leading us to expect an overall positive correlation between fragile codon usage and fragile amino-acid usage . ( ii ) For genes with Ka/Ks close to 0 there is strong selective constraint on the protein sequence and non-synonymous sites are much less modifiable than synonymous sites . Here natural selection can use only one mechanism to increase transcriptional robustness: that of synonymous codon choice since transcriptional robustness can be less readily increased through amino-acid choice . Notably , the protein products of many such genes might be enriched in fragile amino acids due to functional requirements e . g . proteins involved in signal-transduction tend to be tyrosine rich and zinc-finger proteins and proteins enriched in disulfide bonds are cysteine rich . If nonsense errors are deleterious in such genes then there should be strong counterselection of fragile codons to compensate for functionally-determined fragile amino-acid content . For genes under strong selective constraint at the amino-acid level we would therefore expect an overall negative correlation between fragile codon usage and fragile amino-acid usage . Among genes with strong selective constraint on protein sequence , the magnitude of this negative correlation is likely to depend on the abundance of genes with functions requiring high fragile amino-acid content . Considering all human genes ( single-exon and multi-exon genes ) we see a positive correlation between both transcriptional robustness mechanisms ( Spearman correlation for NFCU versus NFAU: rho = 0 . 06 , n = 22985 , p<10−15 ) . More specifically , for human genes having mouse orthologs we can determine how this correlation depends on the strength of selective constraint . We observe that the sign and magnitude of the correlation between both transcriptional robustness mechanisms ( codon-level and protein-level ) is dependent on the strength of selective constraint on protein sequence in agreement with our prediction ( Figure 2 ) . First , for genes towards the upper end of the Ka/Ks range , our first prediction holds . Thus among genes under weaker selective constraint at the protein level ( mean Ka/Ks for the fourth quartile of selective constraint = 0 . 38 ) , we observe a positive correlation between normalized fragile codon usage ( NFCU ) and normalized fragile amino-acid usage ( NFAU ) ( Figure 2 ) . In other words , for genes in which non-synonymous sites are , on average , only 60% less modifiable than synonymous sites , both transcriptional robustness mechanisms are available to selection . Moreover , it is striking that among the top three quartiles of Ka/Ks the positive correlation between both mechanisms increases in magnitude with weakening selective constraint ( Spearman's rho ( p-values ) for NFCU versus NFAU in Ka/Ks quartiles 2-4: 0 . 050 ( p = 0 . 001 ) ; 0 . 068 ( p<10−5 ) ; 0 . 076 ( p<10−6 ) ) . Therefore with increasing ‘flexibility’ of the protein sequence there is an increasing tendency for transcriptional robustness to be realized not only through synonymous codon choice but also through amino-acid choice . However , for genes under strong selective constraint at the protein level ( mean Ka/Ks for the first quartile of selective constraint = 0 . 03 ) the correlation between transcriptional robustness mechanisms is negative ( Spearman's rho for NFCU versus NFAU in Ka/Ks quartile 1: -0 . 025 ( p = 0 . 09 ) ; Figure 2 ) . For such genes non-synonymous sites are , on average , 97% less modifiable than synonymous sites and a requirement for transcriptional robustness can only be accommodated at the level of synonymous codon usage and not at the level of amino-acid usage . In summary , the fact that the genome-wide correlation between usage of fragile codons and fragile amino-acids depends on selective constraint points to a universal requirement for transcriptional robustness to nonsense errors among human genes . A simple interpretation of the lower fragile codon content of single-exon genes is that this observation may be due to differences in splicing-related constraints between single and multi-exon genes . Splicing requires regulatory sequences located in exons known as exonic splicing enhancers ( ESEs ) . Since these hexamer sequences overlap codons , their presence imposes additional constraints on the coding sequence of multi-exon genes [16] in contrast to the coding sequence of single-exon genes that have no such splicing constraints . Therefore , if the nucleotide composition of ESEs is such that they tend to encode fragile codons then a simple difference in ESE density between single and multi-exon genes could provide a trivial explanation for the relative enrichment of fragile codons in the latter group . We found that ESEs indeed tend to encode fragile codons and that ESEs are depleted in single-exon genes . However the difference in fragile codon usage between single and multi-exon genes persists when this is controlled for showing that our observation is not a side-effect of differences in splicing related constraints ( Text S1 , Result B; Figure S1 ) . The patterns of fragile codon usage in single-exon and multi-exon genes could be due to a more familiar source of codon usage bias that has nothing to do with their propensity for nonsense errors . In fly , selection for translational efficiency or accuracy leads to the preferential usage of optimal codons in highly expressed genes [17] . Moreover , recent evidence suggests that such selection might also operate on mammalian genes [18]–[20] . The possible influence of selection for translational efficiency raises two potential concerns for our analysis . First , in human , it might create an artifactual difference in fragile codon usage of the magnitude we observe between single and multi-exon genes leading us to falsely infer a difference in transcriptional robustness between these gene sets . Second , in fly , it might obscure a real difference in transcriptional robustness by homogenizing fragile codon usage between single and multi-exon genes and therefore might invalidate the use of the Drosophila genome as a negative control . Among codons we see no association in either human or fly between fragility with respect to nonsense errors and translational optimality ( see Text S1 , Result C ) . Nevertheless , to account for the possibility that there is an association among genes between transcriptional robustness and selection for translational accuracy we repeated the analysis of fragile codon content in human and fly and controlled for the fraction of optimal codons per gene ( Fop ) [17] , [21] . Comparing human single and multi-exon genes binned in this way revealed that the difference in both FCU and NFCU persists independently of their optimal codon usage ( see Text S1 , Result C; Figure S2 ) . Thus in human , the difference in fragile codon usage between single and multi-exon genes is not an artifact of selection for translationally optimal codon use . Additionally , in fly , the control for optimal codon usage has no influence on the magnitude of the difference between single and multi-exon genes with respect to either FCU or NFCU ( see Text S1 , Result C; Figure S3 ) . Thus in fly , selection for translationally optimal codons does not mute any signal for transcriptional robustness among single-exon genes . The fact that only PTCs lying more than ∼50–55 nts upstream of the last Exon Junction Complex ( EJC ) are thought to be detected by the EJC-dependent NMD pathway suggests that , similarly to single-exon genes , last exons of multi-exon genes constitute a blind-spot for EJC-dependent NMD and are served only by PABP-dependent NMD . Although a mechanism for detection of PTCs downstream of the last EJC has been described ( “fail-safe” NMD ) , its activity is believed to be restricted to exceptional targets [22] . We therefore hypothesized that last exons of multi-exon genes should show a depletion of fragile codons comparable to that seen in single-exon genes . For human multi-exon genes whose coding sequence is completely contained within the last exon we see a significant 8% depletion of fragile codons compared to multi-exon genes having at least two coding exons ( see Text S1 , Result D ) . We next used an intragenic comparison to ask whether the last exons of all human multi-exon genes show a similar depletion of fragile codons . To address this we performed an exon-based analysis of human multi-exon genes that have a reliably annotated last exon and at least one upstream coding exon . We created one group of last exons and one group of upstream exons and compared these with respect to NFCU . We found that fragile codons exhibited a significant ( p = 0 . 0002 , Wilcoxon rank-sum test ) 7% depletion among last exons ( n = 12390 , median NFCU = 0 . 47 ) compared to upstream exons ( n = 112789 , median NFCU = 0 . 50 ) ( Table 4 ) . We repeated the analysis in mouse and observed a 8% depletion of fragile codons ( p = 0 . 002 , Wilcoxon rank-sum test ) among last exons ( n = 13265 , median NFCU = 0 . 46 ) compared to upstream exons ( n = 132678 , median NFCU = 0 . 50 ) . As a control we performed the same analysis on Drosophila multi-exon genes in which NMD should be equally efficient in upstream and last exons . In accordance with our expectation we saw no depletion of fragile codons among last exons ( n = 6304 , median NFCU = 0 . 53 ) compared to upstream exons ( n = 25686 , median NFCU = 0 . 53 ) of Drosophila multi-exon genes . These results were confirmed by an analysis based on matching the last exon of each human multi-exon gene with the upstream sequence of the same gene . There are 12324 human multi-exon genes with reliably annotated last exons and for which NFCU is defined in both the last exon and in the upstream sequence . Among the 12265 genes for which NFCU differs between these two regions , for 6599 ( 54% ) fragile codon content is lower in the last exon than in the upstream sequence whereas in 5666 genes NFCU in the last exon is greater than in the upstream sequence ( p<10−16 , binomial test ) . Considering NFCU for all 12324 pairs of upstream and last exons showed that their fragile codon content is significantly different ( p<10−15 , Wilcoxon signed-rank test ) . Notably , we found that the depletion of fragile codons in last exons of human multi-exon genes can not be explained by the different splicing requirements for the final exon of transcripts [23] ( see Text S1 , Result E ) . The depletion of fragile codons in last exons is particularly striking given the fact that a reduction in the selective costs of premature peptide truncation in last exons might mute any signal of transcriptional robustness caused by inefficient NMD in these exons . In other words , if NMD was equally efficient in upstream and last exons we would expect fragile codons to be enriched in last exons because PTCs generated by mistranscription in last exons are , on average , closer to the normal termination codon ( NTC ) . If these PTCs remain undetected by NMD , the resultant short peptide truncations should have trivial fitness consequences compared to longer peptide truncations caused by undetected PTCs in upstream exons . Consistent with this , the fragile codon content of last exons of human multi-exon genes ( median NFCU = 0 . 47 ) is greater than that of human single-exon genes ( median NFCU = 0 . 43 ) despite the fact that both are invisible to EJC-mediated NMD . We attempted to control for the effect of trivial peptide truncation by analyzing NFCU in regions on either side of the boundary of EJC-dependent NMD but located at least 50 codons away from the normal termination codon . We focused on multi-exon genes having reliably annotated last exons longer than 100 codons and having at least 50 codons in the NMD-competent region of upstream exons ( Figure 3 ) . Specifically , for each multi-exon gene we calculated NFCU in a 3′ window comprising the first 50 codons encoded by the last exon and contrasted this with NFCU calculated from a neighbouring 50-codon 5′ window ending 50 nts upstream of the last exon-exon junction . For human multi-exon genes the 3′ window lies in the NMD-compromised region ( invisible to EJC-dependent NMD ) and the 5′ window lies in the NMD-competent region ( visible to EJC-dependent NMD ) . The boundary between NMD-competent and NMD-compromised regions of human multi-exon genes is described as lying 50–55 nts upstream of the last exon-exon junction ( the “50- to 55- nts rule” ) but the universality of this rule is unclear in the light of more recent genome-wide observations [24] and individual gene studies [25] . Consequently , the parameter settings for positioning these windows were chosen to reflect the uncertainty in the position of this boundary . In this analysis we saw a 10% reduction in fragile codon density in 3′ ( NMD-compromised ) windows compared to 5′ ( NMD-competent ) windows ( n = 2616; 5′ windows , median NFCU = 0 . 50; 3′ windows , median NFCU = 0 . 45; p = 0 . 016 , Wilcoxon rank-sum test ) . As a negative control we repeated this analysis on Drosophila multi-exon genes in which EJC-dependent NMD shows reduced activity implying that surveillance of nonsense errors should be equally efficient in the 5′ and 3′ windows . As expected we saw no difference in fragile codon density between these window sets ( n = 2420; 5′ windows , median NFCU = 0 . 5; 3′ windows , median NFCU = 0 . 5; p = 0 . 199 , Wilcoxon rank-sum test ) . We repeated this test to account for the reduced density of ESEs in human last exons by partitioning the codons in each window into those that overlap ESEs and those external to ESEs and then recalculating NFCU . Both analyses qualitatively agreed with the full analysis considering all 50 codons in each window ( data not shown ) . In summary , we see a consistent pattern of depletion of fragile codons in the last exons of human multi-exon genes when compared with upstream exons . This accords with their status as NMD-compromised regions that are invisible to EJC-dependent NMD and visible only to the PABP-dependent NMD pathway . The selective pressure to avoid fragile codons in last exons imposed by the requirement for transcriptional robustness is likely to be partly offset by increased tolerance of PTCs lying close to the NTC . Nevertheless , transcriptional robustness appears to be an important determinant of codon choice in the last exons of human multi-exon genes as well as in single-exon genes . Since transcriptional robustness is likely to be a consequence of reduced NMD potency we can attempt to dissect the relative contributions to this phenomenon of each of the two mammalian NMD pathways . The large depletion of fragile codons in histone genes relative to other single-exon genes ( Table 1 ) suggests that much of the variation in transcriptional robustness among and within human genes might be explained by variation in the efficiency of PABP-dependent NMD . Notably , the ability of this pathway to detect PTCs increases with the distance between the PTC and the polyA-binding protein ( PABP ) [1] . Single-exon genes produce shorter mRNAs than multi-exon genes ( the coding sequences ( CDS ) of human single-exon genes and multi-exon genes have a median length of 534 nts and 1203 nts , respectively ) . Multi-exon genes may therefore be subject to more potent NMD than single-exon genes owing simply to more efficient surveillance by the PABP-dependent NMD pathway . This follows from the fact that , in multi-exon genes , any PTCs formed by mistranscription will be , on average , more distant from the PABP and thus will elicit PABP-dependent NMD more efficiently . We investigated whether decreased efficiency of PABP-dependent NMD alone is responsible for the greater transcriptional robustness of single-exon genes due to the shorter length of their transcripts . We compared NFCU between single and multi-exon genes binned by CDS length ( Figure 4 ) and found that NFCU was significantly lower in single exon genes than in multi-exon genes for all length bins except the longest ( Q4 ) : the ratios ( p-values; Wilcoxon-rank sum test ) of median NFCU in single-exon genes relative to median NFCU in multi-exon genes for length bins Q1-Q4 are 0 . 92 ( p<10−15 ) , 0 . 90 ( p<10−15 ) , 0 . 97 ( p = 0 . 03 ) , 0 . 99 ( p = 0 . 91 ) , respectively . Thus only the longest single-exon transcripts have fragile codon content equal to that of multi-exon transcripts of similar length . This result suggests that the increased transcriptional robustness of single-exon genes is primarily due to the lack of supplementary nonsense surveillance from EJC-dependent NMD . However , the longest single-exon transcripts also benefit from an increase in the efficiency of PABP-dependent NMD . Repeating this procedure in our analysis of fly genes revealed no difference in NFCU between single-exon and multi-exon genes when length differences are controlled for ( data not shown ) . Thus the modest 2% depletion of fragile codons seen in fly single-exon genes compared to multi-exon genes is entirely due to the inefficiency of PABP-dependent NMD of shorter transcripts ( Table 1 ) . It has recently been demonstrated that splicing enhances NMD in Schizosaccharomyces pombe in a manner apparently independent of the EJC [26] . This organism provides us with an independent test of our hypothesis in a system of splicing-dependent NMD that is mechanistically distinct from that in mammals . We measured NFCU in S . pombe genes and found that single-exon genes ( median NFCU = 0 . 43 ) show a highly significant ( p<10−15; Wilcoxon-rank sum test ) 6% depletion of fragile codons relative to multi-exon genes ( median NFCU = 0 . 46 ) echoing our observations in human and mouse . However , in contrast to the situation in mammals and fly , the CDS of single-exon genes is longer than that of multi-exon genes in S . pombe ( the coding sequence ( CDS ) of S . pombe single-exon genes and multi-exon genes have a median length of 1220 nts and 1053 nts , respectively ) . Therefore we tested for the possibility that the signal of robustness in single-exon genes in S . pombe is muted by the reduced efficiency of PABP-dependent NMD in the shorter mRNAs of multi-exon genes . We compared single and multi-exon genes in bins of equal CDS length and found that fragile codons are significantly depleted in single-exon genes for all length bins except the shortest: the ratios ( p-values; Wilcoxon-rank sum test ) of median NFCU in single-exon genes relative to median NFCU in multi-exon genes for length bins Q1-Q4 are 0 . 99 ( p = 0 . 83 ) , 0 . 92 ( p<10−7 ) , 0 . 93 ( p<10−8 ) , 0 . 95 ( p<10−8 ) , respectively ) . The signal associated with the second mechanism of transcriptional robustness ( amino-acid choice ) is less strong in S . pombe . Among fission yeast proteins we observe a 3% depletion ( p<10−14; Wilcoxon-rank sum test ) of fragile amino-acids in the protein products of single-exon genes ( median NFAU = 0 . 397 ) compared to multi-exon gene products ( median NFAU = 0 . 410 ) . In this study we set out to investigate whether human genes have evolved “transcriptional robustness” to reduce the frequency of mistranscription events leading to nonsense errors . More specifically we hypothesized that such a strategy of nonsense prevention should complement the cure for nonsense errors provided by Nonsense Mediated Decay ( NMD ) . Indeed by comparing single-exon genes ( in which NMD is inefficient ) and multi-exon genes ( in which NMD is much more efficient ) we show that intergenic variation in transcriptional robustness reflects intergenic differences in NMD efficiency . Specifically , the primary hallmark of this robustness is a depletion of “fragile” codons that are susceptible to mistranscription into a STOP codon . We took account of differences between single-exon and multi-exon genes that , although unrelated to transcriptional robustness , might nevertheless covary with fragile codon usage . Our result is not explained by biologically confounding variables such as differences in splicing constraints or transcription associated mutational biases ( see Text S1 , Result F ) nor by possible technical biases due to gene annotation issues or phylogenetic dependencies ( see Text S1 , Result G ) . Having excluded these alternatives we can conclude that , in human , there is a real difference in transcriptional robustness of single-exon and multi-exon genes . Is there an alternative explanation for intergenic variation in transcriptional robustness that is unrelated to the efficiency of NMD ? Before attributing our observation to intergenic differences in NMD efficiency , other differences between single-exon and multi-exon genes that could promote robustness need to be considered . Notably our intergenic analysis makes two assumptions: all genes have ( i ) equal transcriptional error rates and ( ii ) similar selective costs associated with the toxic effects of truncated proteins . However , if either of these quantities was , on average , greater in single-exon genes than in multi-exon genes then this would promote increased transcriptional robustness of single-exon genes even if NMD were equally potent in both groups . Our analysis of intragenic patterns of fragile codon use addresses both assumptions , by isolating the influence on transcriptional robustness of variable NMD efficiency from the influences of variable transcriptional fidelity and selective costs of nonsense errors . First , although transcriptional fidelity might vary between genes , it is not likely to vary within genes . In contrast , the efficiency of NMD varies not only between genes but also within genes since , like single-exon genes , last exons of multi-exon genes are invisible to EJC-dependent NMD in mammals . Therefore if selection for transcriptional robustness was mediated by transcriptional fidelity then this could explain the intergenic but not the intragenic patterns of fragile codon use that we see ( i . e . depletion of fragile codons in last exons ) . Second , although the selective costs of nonsense errors might vary within genes this should result in an enrichment of fragile codons in last exons rather than the depletion observed . Therefore the intragenic pattern of fragile codon usage in mammalian genes , and in particular the depletion of fragile codons in last exons , suggests that variable NMD-efficiency underlies transcriptional robustness . Although these patterns do not implicate variation in transcriptional fidelity or in the selective costs of undetected nonsense errors as unique explanations of transcriptional robustness these factors might also contribute to intergenic differences in robustness . Importantly , the result of our intragenic analysis strongly suggests that constraints on codon usage imposed by the requirement for transcriptional robustness are common to all mammalian genes and are not peculiar to the minority of genes that are intronless in mammals ( ∼11% and ∼15% of genes in the human and mouse genomes , respectively ) . We conclude therefore , that natural selection has promoted a dual strategy of “prevention and cure” to deal with the problem of nonsense transcriptional errors and that these strategies are used interchangeably in mammals . This example of complementarity between strategies for error-prevention and error-mitigation in mammals echoes the recent demonstration in bacteria of complementarity between cis and trans strategies in limiting protein misfolding [27] . There is evidence that both the intron content [5] and the exon-intron structure [24] , [28] , [29] of human genes are shaped by the general mode of action and specific spatial requirements of NMD . We show that variable NMD efficiency also leaves its signature in the coding sequences of human genes and in the amino-acid content of the proteins they encode . This signature can be discerned in patterns of codon choice and amino-acid usage in single-exon genes that together constitute two hallmarks of transcriptional robustness . It is also evident in codon usage in the last exons of multi-exon genes and thus constrains a substantial fraction of sites in human genes . We suggest that fragile codons are counterselected in human genes not because they pose a potential future mutational hazard but because of the immediate hazard associated with undetected nonsense errors during their transcription . However , a notable side-effect of transcriptional robustness is the creation of a “congruent robustness” [30] to future nonsense-creating genetic mutations . The negative selection invoked here is mediated by transcriptional errors and not by genetic mutations . Indeed , transcriptional errors , together with splicing and translation errors , may exert a negative effect on fitness despite the presence of a normal genotype . Such errors in gene expression [31] , [32] may play a much more prominent role in molecular evolution than is currently recognized [33] . One such role may be to expose subtle fitness differences between otherwise equally-fit genotypes , enabling positive selection to explore sequence space in the vicinity of a given genotype by means of a so-called “look-ahead effect” [33] . Equally , we suggest , the evolutionary foresight gained from this effect may reveal pitfalls in sequence space and enable negative selection to purge genotypes that lie close to these pitfalls by providing a preview of their deleterious consequences . This effect provides a rationale for the promotion of robustness by natural selection even when the population genetic conditions of high genetic mutation rate and large effective population size , conventionally thought to be necessary for the evolution of robustness [34] , are not met . Together with the identification of robustness to translational errors [18] and to splicing errors [5] , this study underscores the importance in molecular evolution of the full spectrum of errors made in the decoding of phenotype from genotype . Moreover , the patterns of fragile codon usage and fragile amino-acid usage described for human genes suggest that transcriptional errors are frequent and can be highly deleterious . This raises the question of the past and present impact of such errors on human disease .
We used Ensembl release 49 gene annotations for human and mouse and Flybase release 5 . 4 for Drosophila annotations [35] . For each gene prediction we retrieved its CDS from Ensembl . In the case of multiple alternative transcript predictions , we retained the transcript encoding the longest peptide and used the total number of annotated coding and non-coding exons in the transcript as the exon count for that gene . A dataset of histone genes was constructed by retrieving 72 Genbank accessions for human histone genes from [15] and mapping these to unique Ensembl gene identifiers using the BioMart tool . Fly orthologs of human histone genes were retrieved using BioMart . We constructed a dataset of exonic splicing enhancers ( ESEs ) consisting of 443 hexamer sequences by merging human and mouse ESEs determined using the RESCUE-ESE approach [36]-[38] . We defined two categories of codons using a classification introduced by [14] : “fragile codons” are defined as sense codons that can be converted into a STOP codon by a single point-mutation whereas all other sense codons are defined as “robust” ( Figure 1 ) . The fragile codon usage ( FCU ) metric considers all 61 sense codons and was calculated by enumerating all fragile and robust codons and expressing fragile codon density ( FCU ) as the fraction of all sense codons that are fragile for each gene . Normalized fragile codon usage ( NFCU ) was calculated by considering only groups of codons that are synonymous and have equal GC content but differ with respect to their fragility . These groups were chosen from among codons encoding “facultative amino acids” ( Figure 1 ) . Four such groups have two-members that are respectively fragile and robust: TCA , TCT ( encoding Serine; 1/3 nts are G or C ) , TCG , TCC ( encoding Serine; 2/3 nts are G or C ) , CGA , CGT ( encoding Arginine; 2/3 nts are G or C ) and GGA , GGT ( encoding Glycine; 2/3 nts are G or C ) . A fifth , three membered , group encodes Lysine and consists of TTG , CTT and CTA ( fragile , robust and robust codons respectively; 1/3 nts are G or C ) . For each CDS we computed the fractional fragile codon usage for each of these five synonymous groups ( number of fragile codons/the total number of codons in the group ) . Finally , for each gene we expressed the normalized fragile codon usage ( NFCU ) of its CDS as the average of all fractions that are defined ( i . e . that have a denominator greater than zero ) . For each gene we considered its encoded peptide ( using the longest in the case of alternative isoforms ) and calculated the fraction of all amino acids that are fragile ( Cys , Gln , Glu , Lys , Trp , Tyr ) ( Figure 1 ) considering all amino-acids for each protein ( i . e . FAU = count of fragile amino-acids/total count of amino-acids ) . Normalized fragile amino-acid usage ( NFAU ) was calculated by considering only groups of amino-acids that are encoded by codons of equal GC content . For each gene we considered its longest encoded peptide and , for two separate groups of amino-acids , calculated the fraction of amino-acids that are fragile . The first group consists of four amino-acids encoded by codons for which 1/6 nts are G or C: Tyr ( fragile ) , Lys ( fragile ) , Asn ( robust ) and Phe ( robust ) . The second group consists of eight amino-acids encoded by codons for which 1/2 nts are G or C: Gln ( fragile ) , Glu ( fragile ) , Cys ( fragile ) , Ser ( facultative ) , His ( robust ) , Thr ( robust ) , Val ( robust ) , Asp ( robust ) . For each protein we computed the fractional fragile amino-acid usage for each of these two groups ( count of fragile amino-acids/total count of amino-acids in the group ) . Finally , for each protein we expressed the normalized fragile amino-acid usage ( NFAU ) as the average of defined fractions . Human-mouse orthologs were retrieved using the BioMart tool . In the case of human genes having multiple co-orthologs we retained the longest mouse ortholog ( choosing a random protein in the case of length ties ) . We aligned human-mouse orthologous pairs using CLUSTALW [39] and , using the corresponding CDS , back-translated each alignment to create a codon-based alignment . These alignments were used as input for the yn00 program in the PAML package [40] to estimate Ka/Ks i . e . the ratio of non-synonymous substitutions per non-synonymous site ( Ka ) to synonymous substitutions per synonymous site ( Ks ) . Four groups of human genes of similar selective constraint were constructed based on quartiles of Ka/Ks for all human-mouse orthologs . For each group we calculated the Spearman correlation between normalized fragile codons usage ( NFCU ) and fragile amino-acid usage ( NFAU ) metrics . We performed an exon-based analysis to compare NFCU in last exons and upstream exons . We considered only genes for which we could be sure that the annotated last exon is the true last exon . Last exons were considered as reliably annotated if they ( i ) have a CDS sequence that terminates with a STOP codon , ( ii ) do not have a downstream non-coding exon ( and therefore do not have a downstream EJC ) and ( iii ) have an annotated 3′ UTR of at least 100 nts . For each gene having a reliably annotated last exon we assigned the last exon to one group and each remaining exon to a second group ( “upstream exons” ) . We calculated NFCU for each exon using only those codons completely encoded by the exon and compared NFCU in the “last exon” and “upstream exon” groups . Using the dataset of histone genes retrieved from Ensembl as described above we constructed sequence alignments between all pairs of histone proteins using CLUSTALW [39] and , using the corresponding CDS , back-translated each alignment to create a codon-based alignment . These alignments were used as input for the yn00 program in the PAML package [40] and pairwise sequence divergence was determined by calculating synonymous site divergence ( Ks ) . | In biological systems mistakes are made constantly because the cellular machinery is complex and error-prone . Mistakes are made in copying DNA for transmission to offspring ( “genetic mutations” ) but are much more frequent in the day-to-day task of gene expression . Genetic mutations are heritable and therefore have long been the almost exclusive focus of evolutionary genetics research . In contrast , gene expression errors are not inherited and have tended to be disregarded in evolutionary studies . Here we show how human genes have evolved a mechanism to reduce the occurrence of a specific type of gene expression error—transcriptional errors that create premature STOP codons ( so-called “nonsense errors” ) . Nonsense errors are potentially highly toxic for the cell , so natural selection has evolved a strategy called Nonsense Mediated Decay ( NMD ) to “cure” such errors . However this cure is inefficient . Here we describe how a preventative strategy of “transcriptional robustness” has evolved to decrease the frequency of nonsense errors . Moreover , these “prevention and cure” strategies are used interchangeably—the most transcriptionally robust genes are those for which NMD is most inefficient . Our work implies that gene expression errors play an important role as supporting actors to genetic mutations in molecular evolution , particularly in the evolution of robustness . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"genetic",
"mutation",
"genome",
"evolution",
"population",
"genetics",
"rna",
"stability",
"dna",
"transcription",
"mutation",
"mutation",
"types",
"molecular",
"genetics",
"sequence",
"analysis",
"gene",
"expression",
"gene",
"splicing",
"biology",
"evolutionary",
"ge... | 2011 | Preventing Dangerous Nonsense: Selection for Robustness to Transcriptional Error in Human Genes |
Vast research efforts have been devoted to providing clinical diagnostic markers of myocardial infarction ( MI ) , leading to over one million abstracts associated with “MI” and “Cardiovascular Diseases” in PubMed . Accumulation of the research results imposed a challenge to integrate and interpret these results . To address this problem and better understand how the left ventricle ( LV ) remodels post-MI at both the molecular and cellular levels , we propose here an integrative framework that couples computational methods and experimental data . We selected an initial set of MI-related proteins from published human studies and constructed an MI-specific protein-protein-interaction network ( MIPIN ) . Structural and functional analysis of the MIPIN showed that the post-MI LV exhibited increased representation of proteins involved in transcriptional activity , inflammatory response , and extracellular matrix ( ECM ) remodeling . Known plasma or serum expression changes of the MIPIN proteins in patients with MI were acquired by data mining of the PubMed and UniProt knowledgebase , and served as a training set to predict unlabeled MIPIN protein changes post-MI . The predictions were validated with published results in PubMed , suggesting prognosticative capability of the MIPIN . Further , we established the first knowledge map related to the post-MI response , providing a major step towards enhancing our understanding of molecular interactions specific to MI and linking the molecular interaction , cellular responses , and biological processes to quantify LV remodeling .
Myocardial infarction ( MI ) is a prominent cause of mortality and morbidity worldwide [1] . MI is defined as the death of cardiac myocytes due to prolonged ischemia . As a result of myonecrosis , molecules from injured myocytes are discharged into the blood circulation , and the list of injury markers includes myoglobin , cardiac troponins T and I , creatine kinase-MB , and lactate dehydrogenase [2] . Molecular interactions within the myocardium activate a cascade of cellular responses , including a robust inflammatory response . The cellular responses within the LV are integrated by the extracellular matrix stimuli that bind to surface receptors . As such , the ECM coordinates the healing response to MI [3] , [4] , [5] , [6] , [7] , [8] . Through the last 4 decades , there have been tremendous research efforts towards understanding the immediate myocyte response to ischemia , with the goal of identifying diagnostic indicators as well as targets to preserve myocyte viability . These have resulted in the implementation of several therapeutic strategies , including reperfusion and the use of angiotensin converting enzyme inhibitors [9] , [10] . Currently , 30 day post-MI survival rates approach 90% , and the immediate prognosis is excellent for those patients who receive timely and effective treatment . The number of patients who will go on to develop congestive heart failure , in part as a consequence of this success , however , has increased . While much is known about the events that occur immediately before and after MI , much remains to be mechanistically elucidated regarding the effects of MI on long-term survival . A knowledge map that explores the regulatory relationship among ECM , cellular responses , and biological pathways post-MI is still lacking . Over a million abstracts can be retrieved from PubMed using a keyword search for [“myocardial infarction” or “cardiovascular diseases”] , and massive amounts of genomic and proteomic data and molecular profiles have been deposited in public databases [11] , [12] , [13] . High-throughput protein microarrays have provided efficient procedures to investigate and measure a vast number of protein-ligand interactions in a single experiment . Protein-protein interaction network ( PPI ) analysis using large-scale databases has been one of the most promising computational approaches to integrate experimental data at the molecular and cellular levels [14] , [15] , [16] , [17] . Due to the growing availability of such large-scale datasets , PPIs have been applied to analyze numerous human diseases including lung cancer , breast cancer , and myocardial infarction [18] , [19] , [20] . The reported data which have largely been obtained with different experimental conditions , protocols , species , and research teams are embedded in the literature and distributed in disparate databases . The ability to integrate data from such heterogeneous resources will allow us to extract relevant information and identify knowledge gaps to direct future research efforts . To address these challenges , we report here an integrative computational approach including compiling a MI-specific PPI database through mining PubMed and UniProt to establish a knowledge map for LV remodeling post-MI [21] , [22] . This MI-related knowledge map is the first major step towards enhancing our understanding of molecular interactions specific to MI and linking the molecular interaction , cellular responses , and biological pathways .
MI-related proteins were first obtained from the Online Mendelian Inheritance in Man ( OMIM ) database , PubMed Gene , and PubMed Protein databases by using “myocardial infarction” as the keyword and further refined by our cardiac clinicians ( RAL and RJC ) and cardiac biologist ( MLL ) , producing a list of 38 seed proteins for humans [23] . With these seed proteins and their interacting proteins , we constructed a MI-specific PPI network with a total of 613 proteins ( vertices ) and an associated 4443 interactions ( or edges ) ( Figure 1A ) . Detailed procedures to establish the MIPIN are provided in the Methods . We observed that the MIPIN was strongly connected , in that there was always an edge between any two proteins in the MIPIN . Of the 613 proteins , 70 proteins had only 1 or 2 edges , 121 had 3 to 5 edges , and the rest had >5 edges . The degree distribution of MIPIN closely followed a power law distribution ( Kolmogorov-Smirnoff test , p-value = 0 . 97 , see Methods for details ) , where the degree of a vertex in a network was defined as the number of direct links incident upon that vertex ( Figure 1B ) . The power law distribution indicated that the MIPIN was a scale-free network , which displayed robustness against disruptive failures of random vertices [24] . We performed two statistical tests to evaluate the specificity of the MIPIN . First , interactions were shuffled based on the Erdos-Renyi model , such that the 100 , 000 randomly generated networks each had 613 vertices and 4443 edges , which was the same number as the MIPIN [25] . Compared to the Erdos-Renyi model of random networks , the MIPIN had a lower average value of betweenness centrality while having higher average values of closeness centrality , clustering coefficient , and eccentricity ( empirical p-value<0 . 001 ) , indicating that proteins in the MIPIN were much more closely related to each other than would occur by random chance , and these proteins might have functional relevance . In the second more stringent statistics test , we randomly picked the same number of seed proteins ( n = 38 ) from 14969 human proteins and created 100 , 000 random networks in the same manner we constructed the MIPIN . Each random network had different number of vertices and edges . Compared to the randomly generated networks , the MIPIN had a higher mean value of closeness centrality and eccentricity ( empirical p-value<0 . 05 ) and displayed a distinct distribution of closeness centrality ( Figure 2A ) . We observed a Gaussian-like distribution for closeness centrality in the MIPIN , while closeness centrality distribution in the random networks resembled the Delta function with few vertices having very low value of closeness centrality , regardless of their number of vertices and edges ( Figure S1 ) . We also noticed that the vertices within a small range of degrees in the MIPIN had a larger variance of closeness centrality ( Figure 2B ) , while the closeness centrality remained fairly constant with an increasing number of direct interactions in the random networks ( Figure S2 ) . Figure 2B shows that vertices in the first group [26] displayed substantial differences in closeness centrality with small changes of degree ( natural logarithm of closeness centrality of the red group had a variance of 8 . 25×10−3 ) . On the other hand , as the degree of a vertex increased , the closeness centrality exhibited minor variation ( natural logarithm of closeness centrality of the red group had a variance of 1 . 37×10−3 ) . The overall structure of the MIPIN demonstrated that it was a strongly-connected and scale-free network , indicating that we captured a solid network of protein interactions from the human PPI that was highly specific . Further statistical tests allowed us to evaluate the significance of several MIPIN network properties , including betweenness centrality , closeness centrality , clustering coefficient , and eccentricity . The larger mean values of closeness centrality and eccentricity in MIPIN indicated that the randomly generated networks had more orphan sub-networks in contrast to the single strongly-connected MI network , suggesting proteins in MIPIN were significantly more closely related to each other and have more specific function than would occur by random chance . The localization of MIPIN proteins was determined using Gene Ontology ( GO ) enrichment analysis by DAVID [27] , [28] . GO is a controlled vocabulary of terms that characterizes gene products in terms of their cellular components , biological processes , and molecular functions in a hierarchical structure from the most general to more specialized terms . The cellular components ontology describes locations at the levels of subcellular structures and macromolecular complexes . We focused on classification by cellular components to provide suggestions on the underlying physiological protein functions . More than 65% of the seed proteins were localized in the extracellular region , including vascular endothelial growth factor ( VEGF ) , transforming growth factor beta-1 ( TGFβ1 ) , and tissue inhibitor of metalloproteinase-1 ( TIMP1 ) ( Figure 3 ) . VEGF , TGFβ , and TIMP1 were also localized to platelet alpha-granules that have been known to play an important role in thrombosis , hemostasis , inflammation , atherosclerosis , wound healing , and angiogenesis [29] . In addition , VEGF , TGFβ , and TIMP1 were localized to the ECM , cell surface , and cytoplasmic membrane-bounded vesicle lumens in many cell types , suggesting active roles in multiple pathologies . A list of GO cellular components of the seed proteins were shown in Table S1 . The inclusion of interacting partners of seed proteins in the MIPIN allows us to explore additional potential biomarkers for MI response . These proteins added 57 cellular components to the initial 19 locations ( Figure 4 ) . In addition to the extracellular region , the plasma membrane and cytosol were two preferred sites for most of the proteins in the MIPIN . We also identified a number of macromolecular complexes , including the TGFβ receptor complex , interleukin-1 ( IL1 ) receptor complex , death-inducing signaling complex , origin recognition complex , lipopolysaccharide receptor complex , fibrinogen complex , integrin complex , and transcription factor complex . These complexes strongly suggest the presence of an inflammatory response . The signaling pathway of the lipopolysaccharide receptor complex has been linked to activation and deactivation of macrophages by lipopolysaccharide , a major cell responding to inflammation [30] . Activated macrophages secrete many different inflammatory cytokines , including IL1 and TGFβ . IL1 receptor complex and TGFβ receptor complex are essential factors in the inflammatory response post-MI [31] , [32] . We found 993 enriched GO biological process terms associated with MIPIN using DAVID . To glean functional insight from the large number of enriched GO biological process terms , we adapted a method from Louie et al . to extract the most meaningful biological processes , in terms of specificity [33] . In the GO structural hierarchy , the biological processes can be traversed from the root/parent node ( GO:0008150:“biological process” ) to narrower and more specific definitions in the child nodes , such as from the parent node “regulation of blood coagulation” to its child terms: “positive regulation of blood coagulation” and “negative regulation of blood coagulation” . The function specificity for the GO terms was evaluated based on four measures: number of ancestor terms , offspring score , proportion of terms , and information content . Higher values of these measures indicate higher specificity . A broader , more general term has less number of ancestor terms and more offspring when compared to a narrower , more specific definition . The broadest term “biological process” had no ancestors , since it is the root node in the biological process branch , as the parent of all other GO biological process terms . The offspring score for a GO term was calculated based on the number of offspring for a node such that a higher score represents more specific function . GO proportion described the ratio between numbers of ancestor and offspring terms , with 0 indicating non-specific and 1 indicating the highest specificity . In addition , we considered the probability of observing a GO term because more specific terms annotate less number of genes , and thus were less likely to be found enriched in a dataset . Information content ( IC ) was a normalized score of this probability such that the root node has an IC of 0 , and more specific terms have higher IC . We obtained very different distributions of the 993 biological process GO terms for each of these measures ( Figure 5 ) . The number of ancestors followed a power-law distribution while information content followed a Gaussian-like distribution . These four evaluations illustrated that only a small number of 993 GO terms were specific . Among the most specific GO terms with regards to the number of ancestors , the top 20 terms were related to kinase and transcriptional activities , suggesting the significant signaling in the MIPIN ( Table S2 ) . We obtained 80 enriched GO terms that had only one offspring in the GO dataset while the offspring of the 80 GO terms were not enriched Table S3 ) . These 80 GO terms were the most specific biological processes we could identified for MIPIN . These terms also emphasized the role of kinase signaling , cell apoptosis/necrosis , migration , differentiation , cell-matrix adhesion , ECM remodeling , and inflammatory response . Top 20 GO proportion evaluation resulted in significance of kinase activity and inflammatory responses ( Table S4 ) . The top 20 biological processes with the highest IC score highlighted inflammatory and immune responses ( Table S5 ) . The top two terms in the IC list were “negative regulation of L-glutamate transport” ( p-value<0 . 01 ) and “regulation of L-glutamate transport” ( p-value<0 . 05 ) . Currently , there are very few studies on the role of L-glutamate post-MI . Lofgren et al . found that L-glutamate provides cardioprotection in the same manner as classical ischemic preconditioning [34] . We listed the most significant GO biological process terms based on the four specificity measures and noticed that transcription activity , response to inflammation , and ECM remodeling accounted for the most significant processes ( p-value<0 . 0001 , Table 1 ) . “Positive regulation of JUN kinase activity” ( p-value<0 . 01 ) had the highest GO proportion as of 0 . 987 , the most number of ancestors ( 81 ) and only one child term , and a relatively high IC score as of 7 . 96 , therefore , we identified it as one of the most enriched GO terms in the MIPIN . “Positive regulation of interleukin-6 biosynthetic process” and “positive regulation of interleukin-12 biosynthetic process” ( p-value<0 . 005 ) ranked among the top GO terms with highest number of ancestors , GO proportion and IC score . These two processes represent inflammatory response post-MI . Additionally , three other inflammatory functions “activation of plasma proteins involved in acute inflammatory response” , “connective tissue replacement involved in inflammatory response wound healing” and “wound healing involved in inflammatory response” ( p-value<0 . 0001 ) were ranked high in the top 20 IC list , further confirming the importance of inflammatory response post-MI . These pathways are also important for wound healing . Together with collagen fibril organization and cell-matrix adhesion GO terms , we identified ECM remodeling as another key component post-MI . Based on GO biological process information and MIPIN structure , we predicted protein expression levels in the MIPIN and validated with published results obtained from MI patient data . We automatically text-mined plasma and serum protein expression levels in post-MI patients reported in articles published between Jan 1 , 2005 and May 31 , 2013 . We chose plasma and serum measurements here for an easier clinical study in the future . Abstracts studying association of MI with diabetes , or coronary artery diseases without MI , or protein concentrations being measured after percutaneous coronary intervention post-MI , were not considered . R and Java programs were written to perform XML parsing and text mining on relevant PubMed abstracts ( see Methods ) . From a total of 4326 abstracts , we obtained 21 highly confident up-regulated proteins , and 1 down-regulated protein ( Adiponectin ) , each with expression results confirmed by at least 2 citations ( Table S6 ) . We used a semi-supervised learning method to predict expression changes in other proteins in the network . With the available expression levels on 22 “labeled” proteins as the training set , we predicted 14 up-regulated proteins ( Table 2 ) . To validate the computational predictions , we examined reported literature from 1990 till current and found that 11 of the 14 predicted proteins have supporting experimental evidence . Stromelysin-1 ( matrix metalloproteinase-3 [MMP3] ) , neutrophil elastase ( also named as Human leukocyte elastase , HLE ) , thrombospondin-1 ( TSP1 ) , and fibronectin [35] increased in plasma from patients post-MI [36] , [37] , [38] , [39] , [40] . In mouse models of MI , CD44 increased in LV by 6 hours , C-C motif chemokine 7 ( CCL7 ) increased in ischemic myocardium after 24 hours , ELAV-like protein 1 [41] increased as well as matrilysin ( MMP7 ) [42] , [43] , [44] , [45] . Inhibition of collagen XVIII ( COIA1 ) was found to impair LV remodeling and heart failure in rat MI model [46] . While there was no available expression data on complement factor H ( CFAH ) and matrix metalloproteinase-17 ( MMP17 ) in plasma from patients post-MI , the CFAH polymorphism Y402H has been inversely associated with the risk of coronary heart disease ( CHD ) among women but not men , and MMP17 was found to be overexpressed in atherosclerotic vessels [47] , [48] . We did not find any information regarding TIMP3 , TNF-receptor associated factor 6 ( TRAF6 ) , and brevican core protein ( PGCB ) in the setting of MI either for human or animal studies , although TIMP3 was down-regulated in patients with ischemic cardiomyopathy ( ICM ) and dilated cardiomyopathy ( DCM ) [49] . Further experimental measurements on these proteins are needed to validate our predictions post-MI . The interactions among the 36 proteins were shown in Figure 6 . All 14 predicted proteins and 22 labeled proteins are well connected , except two labeled proteins ( ADIPO and ANFB ) . Since Adiponectin ( ADIPO ) was the only down-regulated protein post-MI , we did not have sufficient evidence to predict other down-regulated proteins . Also , we could not use natriuretic peptides B ( ANFB , also named as BNP for gene name ) to predict any proteins because none of its direct neighbors were connected to proteins with known quantifications , hence having low predictive confidence . Although the GO biological process revealed the overall underlying molecular functions , it could not capture the regulatory dynamics and dependencies required to completely describe a pathway . To have a better understanding of MI pathology , we examined the 613 proteins in the MIPIN and found 48 highly enriched pathways from Biocarta ( http://biocarta . com/; Figure 7 ) . These pathways covered broad categories , including adhesion , apoptosis , cell activation , cell cycle regulation , cell signaling , cytokines/chemokines , developmental biology , expression , hematopoiesis , and immunology . We clustered the 48 enriched Biocarta pathways with respect to their Kappa similarity matrix into 10 functional groups including 4 groups of Kinases Pathways , Angiogenesis , Hypoxia , Acute MI , 2 groups of Inflammatory Responses , LV Remodeling , and other Signaling Pathways ( Figure 8 ) . Each row and column in Figure 8 represented an enriched Biocarta pathway for MIPIN . The sequence of pathways in rows and columns are the same . The row sequence of pathways was shown from the top to the bottom in Figure 8 . Each cell in the figure represented the intersection between a row and a column and the color of a cell represented the similarity between two pathways . The color legend denoted the similarity between two pathways with the red representing high similarity and light color representing low similarity . The strongest similarity was the self-similarity and the color blocks with deepest red color were located on the diagonal of this symmetric figure . It was shown that the acute MI group ( block AMI ) shared high similarity within the block and relative low similarity with only two pathways h_sppaPathway in block angionenesis ( block A ) and h_p53hypoxiaPathway in block hypoxia ( block H ) . h_sppaPathway denoted “aspirin blocks signaling pathway involved in platelet activation” and h_p53hypoxiaPathway denoted the role of p53 and hypoxia in the cardiovascular system . Interestingly , by checking the color of the intersections of h_p53hypoxiaPathway and h_sppaPathway , the similarity between these two pathways were very low , suggesting no proteins in common in these two pathways and these two pathways could independently contribute to acute MI . Kinases ( KP ) and signaling pathway ( SP ) blocks shared high similarity with more pathways in general since they transmitted spatial signals to trigger pathways related to cellular functions , which was illustrated by the appearance of light yellow boxes in the rows/columns representing KP and SP blocks . Specifically , kinases pathway blocks KP1 , KP2 , and signal transduction pathway SP were closely related to inflammatory response IR1 . Kinases pathway block KP3 was closely related to hypoxia block H . Kinases pathways block KP4 was closely related to angiogenesis block . As an example , platelet activation ( h_sppaPathway ) was one of the pathways that shared similarity with the most number of pathways Figure 8 . It shared higher similarity with kinases pathway block KP1 and low similarity with inflammatory response block IR1 ( as shown in the 3rd column from the right or 3rd row from the bottom ) . Meanwhile , KP1 and IR1 shared high similarity , suggesting a cause-effect relationship from platelet activation , kinases pathway KP1 to inflammatory response IR1 cascade . Platelet activation pathway also shared high similarity with KP3 , KP4 , and angiogenesis ( A ) blocks , suggesting a possible regulation between platelet activation and angiogenesis . Although there was no specific pathway named LV remodeling in Biocarta , we defined the Inhibition of Matrix Metalloproteinases pathway ( h_reckPathway ) as part of LV remodeling in our knowledge map since the pathway was closely related to ECM degradation . There are 9 proteins listed in pathway by Biocarta , including MMP-2 , -9 , TIMP-1 , -2 , -3 , -4 , reversion-inducing-cysteine-rich protein with kazal motifs ( RECK ) , v-Ha-ras Harvey rat sarcoma viral oncogene homolog ( RAS ) and all of them were included in our MIPIN . This pathway did not show high similarity with any other pathways in Figure 8 though illustrating low similarity with h_pmlPathway in KP1 block , h_bcrPathway and h_pyk2Pathway in KP4 block , and 7 pathways in angiogenesis block , suggesting possible regulation among LV remodeling , inflammatory response , and angiogenesis . To better understand Figure 8 , pathways clustered in each functional group were listed in Table 3 , and 160 proteins with specific regulatory relationship in each functional group were listed in Table S8 . This forms the basic knowledge map for MI response that links proteins to specific pathways and functional groups . Combining functional information for all 613 potential MI related proteins extracted by MIPIN , including cellular components , biological processes , and specific pathways , we established the knowledge map for MI ( Figure S3 ) . Essentially , the knowledge map summarizes important spatial and temporal aspects of the static MIPIN; it describes the progression of MI and involvement of different proteins in three major phases: Development of MI ( hypoxia and acute MI ) , response to MI ( signaling pathway , kinases pathway , and inflammatory responses ) , and tissue remodeling ( left ventricle remodeling and angiogenesis ) .
The goal of this study was to establish a framework to 1 ) automatically extract the information embedded in MI-related PubMed abstracts and reported data through a PPI network , 2 ) integrate the information into a knowledge map for MI response , and 3 ) cluster proteins in the knowledge map based on their functions . In this study , we started from the seed proteins for MI and PPI databases at molecular level , extended to cellular components of the proteins at cellular level , and further mapped the information to functional responses and specific pathways to illustrate a complete framework that integrates molecular , cellular , and functional analysis . There are three major contributions of this study . First , we established a MI-specific PPI network and confirmed its specificity with two different statistical analyses . We predicted expression levels of 14 proteins in the MIPIN based on the up/down regulations of 22 proteins . The predicted protein expressions from computational analyses agreed well with reported experimental measurements . Second , we illustrated the importance of inflammatory and ECM remodeling responses in LV remodeling post-MI . Most proteins in the MIPIN were localized primarily in the extracellular regions and the plasma membrane . Additionally , transcription activity , ECM remodeling , and inflammatory response were the main functional themes of the MIPIN . In fact , almost half of the 22 highly confident proteins were inflammatory or extracellular proteins , demonstrating that these two phases are very crucial in determining the outcome of MI . Third , we established the first knowledge map for MI response based on the clustered pathways . This is the first knowledge map constructed by integrating our knowledge obtained from molecular , cellular , and functional factors via PPI , cellular components , biological processes and pathways . In addition , the knowledge map illustrated the temporal response from development of MI to tissue remodeling and the related proteins at each stage . The approach to establish the knowledge map for MI could also be applied to other diseases . Our results illustrated that using the structural property of the PPI network is a promising technique to distinguish functional specific networks from random networks . However , individual structure property alone may not be sufficient to identify significant markers . Degree centrality provides independent evaluation of direct links of a vertex . Intuitively , a hub protein with higher degree may represent a significant marker . However , this cannot be confirmed with current clinical practice . For example , cardiac troponin I ( cTnI ) is a well-known biomarker for MI but cTnI only has a degree of 3 in our network [50] . Additionally , MMP9 and TIMP1 have been reported as key regulators of LV remodeling post-MI in a number of publications , while MMP9 had a degree of 36 and TIMP1 had 12 , the average degree of MIPIN was 15 [51] , [52] . Another structure property , betweenness , denotes how frequently a vertex or edge is used while walking through the network with shortest path . The combination of different structural properties might be a promising way to identify key markers . For example , a vertex with small degree and high betweenness denotes a protein that is frequently used to transmit information in the network , suggesting its significance as a bottle neck of the network or cross talk between biological processes . More accurate analysis of such evaluation scheme will be conducted in our future research . Our results highlight the influence of the early inflammatory response initiated after tissue hypoxia . Following hypoxia , up-regulation of RAS , focal adhesion kinase 1 ( FADK1 ) , paxillin ( PXN ) , and p53 simultaneously induce at least four major cellular activities , including cell proliferation , migration , apoptosis and necrosis . Proliferation of endothelial cells increases the production of nitric oxide ( NO ) , which plays an important role in the later phase of LV remodeling and wound healing . Fibroblasts and myofibroblasts deposit a network of collagen at the infarct site , preparing for the formation of tissue granulation . Collectively , cell proliferation , migration , apoptosis and necrosis contribute to angiogenesis parallel to scar formation . In summary , we report here the establishment of the first MI-specific PPI network that can be used as a foundation to interrogate the literature for candidate biomarkers of adverse remodeling post-MI .
In order to acquire a list of proteins related to MI , we initiated a keyword search for “myocardial infarction” in three different databases including OMIM , PubMed Gene and PubMed Protein , resulting in an initial pool of 658 genes from OMIM and PubMed Gene and 2319 protein sequences from PubMed Protein databases . Because the obtained genes were retrieved using both animal and clinical studies , all the genes and proteins retrieved from OMIM , PubMed Gene , and PubMed Protein databases were matched for human protein names in UniProt , yielding 709 proteins ( Table S9 ) . By evaluating the description of the genes obtained from OMIM , terms not related to MI response were revealed ( e . g . , stroke , arrhythmogenic , cardiomyopathy , and arterial calcification ) . These genes were removed from our list . We also removed proteins directly related to myocytes , since these proteins reflect more the pre-MI or acute MI instead of post-MI response . From this , we were left with 22 MI response related genes . Searching PubMed Gene and Protein databases provides a candidate list of genes and protein sequences potentially associated with MI; however , this search strategy does not provide any description of the retrieved genes and proteins . We verified additional 16 seed proteins associated with MI using genome wide disease association databases , GENERIF and PubMed . This led to a total of 38 seed proteins including the major ones previously identified in our experiments , including collagen , MMP9 , TIMP1 , TNFα , TGFβ , and monocytes chemotactic protein-1 ( MCP1 ) . All seed proteins were associated with MI in at least 2 independent manuscripts , as shown in Table 4 . Consistent with a strong role in the wound healing response , a significant portion of the seed proteins were localized to the ECM . To verify whether our selection of seed proteins was biased , we checked cellular localization of all MI related proteins obtained from OMIM , PubMed Gene , and PubMed Protein databases and encountered a similar result; most of the proteins were localized in the extracellular region and plasma membrane ( Table S10 ) . These results indicate that ECM proteins are more likely play a key role in MI response and suggest that our seed protein selection was not biased . From the seed protein list , we searched for all proteins interacting with seed proteins and interactions among the extended proteins through ConsensusPathDB-human , which integrates protein-protein interactions in Homo sapiens from different databases such as Intact , DIP , MINT , HPRD , BioGRID and MIPS [22] . Subsequently , we constructed the MIPIN using ‘igraph’ in R [53] . Each vertex of the network represents a protein and each edge between two vertices represents a protein-protein interaction . The resulting MIPIN consists of 613 vertices and 4443 edges . The degree distribution of MIPIN was examined by the procedure proposed by Clauset et al . and implemented in R [54] . Parameters were estimated based on the theoretical cumulative distribution , where x , in this case , was degrees of MIPIN vertices . The degree distribution was fitted with xmin = 31 and α = 3 . 52 ( Kolmogorov-Smirnoff test , p-value = 0 . 97 ) . Additionally , the Kolmogorov-Smirnoff test was performed to examine how well the estimated power law distribution fitted MIPIN vertex degrees . If the Kolmogorov-Smirnoff p-value<0 . 05 , we reject the hypothesis that the original data is drawn from the fitted power-law distribution . Otherwise , the higher the Kolmogorov-Smirnoff p-value is above 0 . 05 , the better the estimated power-law distribution fits the data . There were several different measures used to characterize the properties of the network , including betweenness centrality , closeness centrality , clustering coefficient , degree centrality , eccentricity , and graph density . The betweenness centrality characterizes the direct and indirect influences of vertices at distant network sites [55] . Closeness centrality measures how many steps are required to access every other vertex from a given vertex [55] . The vertex with the largest value of closeness centrality performs the least amount of steps to sequentially spread information to other reachable vertices from that vertex in the network . Clustering coefficient describes the connectivity of the neighborhood of a vertex [56] . Higher clustering coefficient means more neighbors are connected to each other . Eccentricity of a vertex measured the shortest path distance from the farthest vertex in the graph [57] . We compared the value of six aforementioned measures of MIPIN with the average measurements of randomly generated networks . The empirical p-values for each measure were then calculated by counting the number of random networks whose average measures were equal to , greater or smaller than the corresponding values from MIPIN . We examined the functional organization of MIPIN with enriched GO terms using DAVID Functional Annotation Tool [27] . In DAVID , we set the count to be 2 and 0 . 05 for EASE , a modified Fisher Exact P-Value . We further adapted the method proposed by Louie et al . to measure the specificity of the enriched GO terms for the MIPIN [33] . We computed four measurements to describe the function specificity of enriched GO term lists . We searched the key word “ ( myocardial infarction ) AND ( plasma OR serum ) ” on PubMed with “Homo Sapiens” as species from Jan 1 , 2005 until May 31 , 2013 . This search resulted in 4326 abstracts . To reduce laborious manual effort , we developed a data mining program written in R using available XML parser and text mining software [58] , [59] . The program required two input files , a list of protein aliases and a dictionary of words . We took advantage of a feature offered by UniProt in which users can submit a list of proteins and receive their full names and aliases in XML format . In order to obtain the full names and aliases of MIPIN proteins , we wrote a Java program to parse downloaded UniProt XML files and extract relevant information . The Java program can also be used to retrieve other protein features such as protein structures , domain , and citations in PubMed . The dictionary of words contained commonly used word indicating protein changes such as , “elevate” or “up-regulate” for positive change “UP” , or “down-regulate” or “inhibit” for negative change “DOWN” ( Table S7 ) . Abstracts in “txt” format were initially broken into separate sentences . If words of change and names of any proteins were found in the same sentence , we recorded the protein names with the associated words , and PubMed ID of the abstracts . The final output was manually checked to ensure complete and accurate reporting of available protein concentrations . This program significantly reduced the reading time of 4326 abstracts to extracted key sentences . From these abstracts , we retrieved a small number of proteins with quantified concentrations in plasma or serum post-MI and assigned as labeled proteins in the MIPIN . A large number of MIPIN proteins did not have quantified concentrations and were assigned as unlabeled proteins . We applied semi-supervised learning to predict unlabeled proteins with the labeled protein set . The key component of this method is defined in the similarity matrix . The similarity matrix represents pair-wise similarity or dissimilarity between pairs of vertices . In this case , we combined graph structure similarity matrix evaluated using Jaccard coefficients and functional similarity matrix evaluated using Wang's method [60] . The Jaccard similarity matrix J of a graph G is a |V ( G ) |×|V ( G ) | square matrix , where |V ( G ) | denotes the number of vertices in the graph G . The Jaccard similarity coefficient of two vertices/proteins i and j was defined as , ( 4 ) where Ni and Nj represented the set of direct neighbors of vertex i and j , respectively [61] . It follows that the diagonal of matrix J is 1 . Besides structural information embedded in the Jaccard similarity matrix , we also integrated biological functions obtained from GO terms by calculating GO biological process similarity matrix GS also of size |V ( G ) |×|V ( G ) | . The pairwise functional similarity between protein i , annotated by GO biological process term sets GOBPi = ( gobpi1 , gobpi2 , … , gobpim ) , and protein j , annotated by GO biological process term sets GOBPj = ( gobpj1 , gobpj2 , … , gobpjn ) , is defined as , ( 5 ) where Sim ( gobpi , GOBPj ) was defined as the maximum semantic similarity between term gobpi and any of the terms in set GOBPj , with m and n represented terms in the ith and jth GOBP term sets , respectively [60] . The semantic similarity between a pair of GO terms can be determined based on their locations in the directed acyclic GO graph and their semantic relations , which can be ‘is-a’ or ‘part-of’ , with their ancestor terms . The GS matrix is symmetric . We chose the Wang method , because the measurement algorithm offered two advantages . First , it only depends on the relationship of the GO terms within a specific ontology , which is the biological process in this case . Second , it avoids the effect of shallow annotation on the semantic relationships between child and parent terms ( i . e . , with the same parent , a pair of terms near the root should have larger semantic differences than a pair of terms far away from the root ) . Thus , the algorithm provided a consistent semantic similarity measurement between a pair of GO terms . We combined Jaccard similarity matrix J and GO biological process similarity matrix GS to produce the final similarity matrix W . The ij element of final similarity matrix W was defined as , ( 6 ) Let L denote the labeled proteins and U denote the unlabeled proteins . The similarity matrix W could be partitioned as ( 7 ) Let where D was the diagonal row sum matrix of W , and was a binary vector describing the concentrations of labeled proteins post-MI with 1 for positive change “UP” and 0 for negative change “DOWN” . Then the predicted concentration vector can be computed using the fits algorithm , ( 8 ) The predicted concentrations were further updated with the sequential predictions algorithm to drive the estimates towards global point estimates . The algorithm ranked the unlabeled data into k number of regions , such that the unlabeled set connecting to the most number of labeled proteins was employed first with the fits algorithm , and penalized unlabeled proteins farther away from labeled proteins with inverse regularization penalty l . It was reasonable to initialize the fits algorithm with the protein having the highest labeled connectivity , and repeat with each subsequently ranked protein . We assigned the number of regions k to be the number of unlabeled proteins . Since we wanted to maintain a moderate regularization , the inverse regularization penalty l was set to be 2 . The prediction process was implemented with the package ‘spa’ in R [62] . A total of 48 enriched Biocarta pathways were retrieved from DAVID using 613 proteins in MIPIN with ‘Count’ set to be 2 and EASE set to be 0 . 05 . The relationships between proteins and associated pathways could be simplified to a binary matrix of M rows and n columns , where M was the number of enriched pathways and n was the total number of associated proteins with enriched pathways ( Table 5 ) . If a protein was involved in a pathway , the corresponding score was denoted as 1 , otherwise 0 . Based on the pathway matrix , we used Kappa statistics to evaluate pathway pairwise similarity matrix based on the belief that pathways sharing common proteins might be related to one another [63] . Considering two pathways I and J ( I≠J; I , J = 1 , 2 , … , M ) , we could determine the number of proteins annotated by both pathways , the number of proteins annotated by pathway J but not I , the number of proteins annotated by pathway I but not I , and the number of proteins not annotated by neither pathway among the union of proteins annotated by all pathways , denoted as a , b , c and d , respectively . Kappa score κ was defined as where Pr ( agree ) was the observed percentage agreement and Pr ( random ) was the overall probability of random agreement . A high Kappa score indicated that two pathways share many common proteins and vice versa . The observed percentage agreement Pr ( agree ) could be calculated as , ( 9 ) To calculated the overall probability of random agreement Pr ( random ) , we noted that pathway α annotates and pathway β annotates of total associated proteins . Thus , the probability that both pathways randomly annotate the same proteins was and the probability that neither pathway randomly annotate the same proteins was Thus , the overall probability of random agreement Pr ( random ) could be calculated as , ( 10 ) Kappa score κ could be rewritten as , ( 11 ) | Heart attack , known medically as myocardial infarction , often occurs as a result of partial shortage of blood supply to a portion of the heart , leading to the death of heart muscle cells . Following myocardial infarction , complications might arise , including arrhythmia , myocardial rupture , left ventricular dysfunction , and heart failure . Although myocardial infarction can be quickly diagnosed using a various number of tests , including blood tests and electrocardiography , there have been no available prognostic tests to predict the long-term outcome in response to myocardial infarction . Here , we present a framework to analyze how the left ventricle responds to myocardial infarction by combining protein interactome and experimental results retrieved from published human studies . The framework organized current understanding of molecular interactions specific to myocardial infarction , cellular responses , and biological processes to quantify left ventricular remodeling process . Specifically , our knowledge map showed that transcriptional activity , inflammatory response , and extracellular matrix remodeling are the main functional themes post myocardial infarction . In addition , text analytics of relevant abstracts revealed differentiated protein expressions in plasma or serum expressions from patients with myocardial infarction . Using this data , we predicted expression levels of other proteins following myocardial infarction . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"systems",
"biology",
"text",
"mining",
"natural",
"language",
"processing",
"signaling",
"networks",
"biology",
"computational",
"biology"
] | 2014 | Integrative Computational and Experimental Approaches to Establish a Post-Myocardial Infarction Knowledge Map |
In neurons , the timely and accurate expression of genes in response to synaptic activity relies on the interplay between epigenetic modifications of histones , recruitment of regulatory proteins to chromatin and changes to nuclear structure . To identify genes and regulatory elements responsive to synaptic activation in vivo , we performed a genome-wide ChIPseq analysis of acetylated histone H3 using somatosensory cortex of mice exposed to novel enriched environmental ( NEE ) conditions . We discovered that Short Interspersed Elements ( SINEs ) located distal to promoters of activity-dependent genes became acetylated following exposure to NEE and were bound by the general transcription factor TFIIIC . Importantly , under depolarizing conditions , inducible genes relocated to transcription factories ( TFs ) , and this event was controlled by TFIIIC . Silencing of the TFIIIC subunit Gtf3c5 in non-stimulated neurons induced uncontrolled relocation to TFs and transcription of activity-dependent genes . Remarkably , in cortical neurons , silencing of Gtf3c5 mimicked the effects of chronic depolarization , inducing a dramatic increase of both dendritic length and branching . These findings reveal a novel and essential regulatory function of both SINEs and TFIIIC in mediating gene relocation and transcription . They also suggest that TFIIIC may regulate the rearrangement of nuclear architecture , allowing the coordinated expression of activity-dependent neuronal genes .
The adaptation of living organisms to their surroundings depends on their ability to fine-tune their behaviors in response to novel conditions . Failure to adapt rapidly to environmental changes compromises behavioral responses such as memory formation , which are essential for survival . At a cellular level , exposure to environmental enrichment correlates with a number of morphological changes , ranging from increased dendritic growth and branching to enhanced synaptogenesis and hippocampal neurogenesis [1] . Many of these enrichment-mediated cellular changes depend on the expression of specific genes involved in neuronal plasticity , such as the neurotrophins Brain Derived Neurotrophic Factor ( BDNF ) and Nerve Growth Factor ( NGF ) , as well as synaptic proteins , including PSD95 and glutamate receptor subunits [2]–[4] . In neurons , transcriptional activation depends on a host of molecular events that act on at least three distinct , yet interconnected levels . The best-characterized mechanism relies on the interaction of nuclear factors with specific regulatory elements within gene promoters . Binding of co-repressor complexes often containing histone deacetylases , inhibits transcription by inducing chromatin modifications that prevent the recruitment and assembly of RNA polymerase II ( RNAPII ) complexes [5] . Conversely , interaction of co-activators with specific DNA sequences within gene promoters is associated with stimulus-dependent transcription [6] . Transcriptional activators and repressors are often simultaneously detected on promoters of both active and inactive genes [7] , suggesting that a dynamic balance between gene activation and inhibition may determine the transcriptional outcome . A second level of regulation relies on epigenetic modifications of histone proteins , including acetylation , methylation and phosphorylation . Post-translational modifications of histones and DNA methylation induce structural changes of the chromatin , and provide docking sites for the recruitment of transcriptional cofactors [6] , [8] , [9] . Importantly , chromatin modifications contribute to determining whether genes are silenced , expressed or maintained transcriptionally inactive , yet poised for activation . Promoters of poised genes are characterized by stalled RNAPII [10] , [11] and although they share several epigenetic marks with actively transcribed genes , they become expressed only upon stimulation [12] , [13] . This mechanism is especially relevant for genes that undergo very rapid stimulus-dependent transcriptional activation , such as immediate-early genes ( IEGs ) [14] . Both binding of transcription factors and epigenetic modifications associated with transcriptional activation are also observed at extragenic regions that function as enhancers [15] , [16] . Finally a third level of transcriptional regulation depends on the three-dimensional organization and functional compartmentalization of the nucleus [17]–[19] . In several mammalian cell types , transcription of genes that are concomitantly activated in response to extracellular stimuli or during cell differentiation occurs specifically at intranuclear foci enriched with active RNAPII [20] , [21] . These transcriptional hubs are known as transcription factories ( TFs ) and in addition to RNAPII , they often include transcription factors [19] , [22] . To investigate how synaptic activity regulates gene transcription in vivo , we performed a genome-wide ChIPseq analysis of histone H3 acetylation at lysine residues 9 and 14 ( H3K9K14ac ) using somatosensory cortex of mice exposed to novel enriched environmental ( NEE ) conditions . De novo acetylated regions contained Short Interspersed Elements ( SINEs ) that interacted with the general transcription factor TFIIIC . SINEs are generated by retrotransposition of genes transcribed by RNA polymerase III ( RNAPIII ) and , although they are present within the genome in hundreds of thousands of copies , their function is only partially understood [23] . TFIIIC is part of the RNAPIII complex and mediates both the recognition and the binding of RNAPIII to gene promoters [24] . Interestingly , in mammalian cells , ChIPseq analyses of TFIIIC occupation demonstrated a widespread binding of TFIIIC across the genome in the absence of RNAPIII [25]–[27] , suggesting that its role is not limited to mediating RNAPIII-dependent transcription . Here , we show that in neurons , TFIIIC regulates activity-dependent relocation to TFs and transcription of RNAPII-dependent gene loci . Inhibition of TFIIIC levels dramatically increased both dendritic growth and branching . The eukaryote genome harbours hundreds of TFIIIC binding sites with unknown function . Our findings indicate that , in neurons , TFIIIC mediates the rearrangement of nuclear architecture , possibly by coordinating the simultaneous expression of activity-dependent neuronal genes necessary for dendritic growth .
Exposure to NEE represents a physiological mode of neuronal stimulation that combines sensory , cognitive and social stimuli [1] , [28] . In rodents , NEE elicits a variety of plastic responses , including increased dendritic arborization , synaptic density , neurogenesis , and improved memory functions . Moreover , in humans , exposure to an enriched environment has a beneficial effect on many pathological conditions , contributing to ameliorate symptoms associated with several neurodegenerative disorders [1] , [29] . The mechanisms underlying these changes are not yet fully elucidated , however enhanced synaptic activity and neurotrophin signaling are implicated [29] . We first studied whether neuronal activation controls gene expression in vivo by exposing adult mice to NEE for 45 minutes ( Figure S1A ) . As a control , mice were maintained in standard cages for the same amount of time . A short-term exposure was chosen in order to characterise the early transcriptional events that regulate activity-dependent neuronal functions [30] . In situ hybridization of mouse somatosensory cortices demonstrated that a short-term exposure to NEE was sufficient to elicit a robust expression of many activity-dependent genes , including c-Fos [31] and Arc [32] , [33] both in the cortex and hippocampus ( Figure S1B–C ) . To characterize the epigenetic modifications associated with the transcriptional program activated by NEE , we employed ChIPseq assay , a technique that combines chromatin immunoprecipitation with large-scale direct ultrahigh-throughput DNA sequencing . Various posttranslational modifications of histones have been associated with “open” chromatin and gene expression , including histone H3 acetylation at lysine residues 9 and 14 , and histone H3 methylation on lysine 4 . H3K9K14ac represents an epigenetic mark , which is particularly useful in the identification of active neuronal genes , as this chromatin modification has been detected on both promoters and extragenic enhancers of actively transcribed genes in cortical neurons [15] , [34] . Mice were either exposed to NEE conditions or maintained in standard cages , the somatosensory cortex was rapidly dissected and regions differentially acetylated on H3K9K14 in response to synaptic activity were analysed by ChIPseq . At least 3 . 5 millions unique reads were obtained from control and NEE-exposed ChIPseq libraries ( Table S1 ) . Differentially acetylated genomic regions were identified by counting all reads mapping within 2 kb sliding windows across the genome and the differential enrichment relative to a Binomial null model was analysed ( see the Materials and Methods section for details ) . This approach provided statistical robustness despite the relatively low number of sequenced reads . 9 , 811 regions ranging from 2 to 8 . 5 kbp in size showed increased levels of H3K9K14ac ( referred here as +Δac ) , whereas 13 , 158 regions , ranging from 2 to 11 . 5 kbp had decreased H3K9K14ac ( −Δac; detailed results of the ChiPseq screens are provided in Dataset S1 ) . Most regions that presented activity-dependent changes of H3K9K14ac levels encompassed the body of annotated genes ( Figure S1D ) whereas 895 +Δac and 1 , 263 −Δac regions overlapped with transcription start sites ( TSS ) . Next , we analysed the transcriptional response to NEE in the somatosensory cortex of mice that were exposed to enriched environmental conditions for 3 hours or maintained in standard cages . Genome-wide microarray analysis revealed that 11 , 071 genes were expressed in both control and NEE-exposed mice ( henceforth referred to as Constitutively Expressed , CE ) while 17 , 451 remained transcriptionally silent ( Constitutively Silent , CS ) . Among the transcripts that were differentially expressed , 196 were induced by at least 1 . 24 fold in response to NEE ( NEE-induced , NI ) whereas 70 were repressed ( NEE-repressed , NR; Table S2 ) . Transcriptional activation of Arc , c-Fos and Gadd45b in response to NEE was further validated by qRT-PCR analysis ( Figure S1C ) . Surprisingly , among the 1 , 663 genes showing NEE-dependent increase of H3K9K14ac at the TSS , only 28 were also transcriptionally activated ( Figure S1E ) . A possible explanation for this finding is that a more prolonged stimulation may be necessary in order to trigger the transcription of most de novo acetylated genes . To further characterize the correlation between gene promoters acetylation and transcription , H3K9K14ac profiles at TSSs of constitutive ( CE ) , silent ( CS ) , NEE-induced ( NI ) and NEE-repressed ( NR ) genes were analyzed before and after NEE exposure ( Figure 1A–D and Table S2 ) . In control conditions , H3K9K14ac tag density analysis showed an enrichment of at least 2-fold at CE gene promoters , when compared to CS ( Figure 1A , B ) . H3K9K14ac profiles of both NI and NR gene pools almost perfectly overlapped with CE genes , indicating that acetylation levels of promoters undergoing rapid changes of transcriptional activity are remarkably stable . Surprisingly , promoters of NI genes were already acetylated prior to exposure to NEE ( Figure 1A , B ) , suggesting that , in unstimulated somatosensory cortex , promoters of activity-dependent genes are hyperacetylated and poised for transcription . A recent study has shown that in neurons maintained in resting conditions , the binding of RNAPII to promoters of many activity-dependent IEGs was comparable to depolarized neurons . Interestingly , RNAPII stalling was necessary for fast transcriptional activation of IEGs [14] . It is therefore conceivable that in unstimulated neurons , the combination of promoter acetylation on H3K9K14 and RNAPII binding represents a feature of rapidly inducible genes . The finding that promoters of both stably expressed and inducible genes shared a virtually indistinguishable pattern of H3K9K14ac , prompted us to investigate whether acetylation of chromatin regions other than TSSs were uniquely associated with NEE-activated genes . To identify putative regulatory elements that mediated transcriptional activation of inducible genes , we employed the motif-prediction tool NestedMICA [35] and analysed the +Δac regions identified by ChIPseq in the cortex of mice exposed to NEE . Motif inference identified two elements of 20 and 29 bp ( referred here as Ac1 and Ac2 ) that were over-represented in +Δac regions ( Figure 1E ) . Further investigation revealed that both motifs overlapped with distinct portions of B1 SINEs ( Figure S1F ) . SINEs , together with the Long Interspersed Elements ( LINEs ) and the Long Terminal Repeats ( LTRs ) are among the most abundant classes of retrotransposons present in the mammalian genome . LINE-mediated retrotransposition has been demonstrated to occur at a much higher rate than previously estimated , thereby contributing to generate cell diversity . In neurons , L1 retrotransposition alters the expression of many neuronal genes , influencing cell fate in vitro and inducing somatic mosaicism in vivo [36] , [37] . Because SINEs are short elements with no coding capacity , they show hardly any retrotransposition activity and have long been considered as “junk DNA” . However , a number of studies demonstrated that they possess diverse and evolutionarily important biological functions , from the regulation of transcription to the targeting of mRNAs [38] . Recent studies have also demonstrated that SINEs may influence the development of the nervous system . AmnSINEs , a highly conserved family of SINEs identified in the genome of amniota , possess enhancer properties and contribute to the expression of genes necessary for brain development . For example , amnSINEs proximal to Fgf8 and Satb2 genes , possibly acting as tissue-specific enhancers of transcription , recapitulate the expression pattern of Fgf8 and Satb2 in the developing forebrain [39] , [40] . SINEs are characterised by an RNAPIII promoter encoding two elements known as A and B boxes , although within SINE families , variable levels of conservation are observed [24] , [41] . The motif Ac2 that we identified includes 15 bp of the 16 bp long consensus sequence of the B box ( Figure S1F ) . Despite the fact that Ac1 and Ac2 elements specifically identified B1 SINEs , the occurrence of all SINE families within +Δac regions was also determined . Alu ( or B1 ) , B4 and MIR SINEs displayed a significantly higher frequency within +Δac regions when compared with randomly selected genomic sequences of comparable size ( Figure S1G ) . Conversely LINEs showed a lower frequency within +Δac regions . At least one SINE was present within 73% of 9 , 811 +Δac regions . Given that many sequenced tags derived from repetitive elements ( such as SINEs ) cannot be univocally mapped and were therefore excluded from the ChIPseq analysis , it is likely that these data underestimate the frequency of SINEs within +Δac regions . Because +Δac regions are located mostly within gene loci ( Figure S1D ) , and gene-rich regions are characterized by a higher presence of SINEs [42] , the increased SINEs frequency in +Δac regions shown in Figure S1G may simply reflect the genomic context of the acetylation changes . To exclude this possibility , the relative distribution of de novo acetylated SINEs within the genomic context was analyzed using genes grouped accordingly to their transcriptional response to NEE . Despite having stable and comparable levels of acetylation at the TSS ( Figure 1A–D ) , NI genes showed a higher frequency of acetylated SINEs both upstream ( −95/−20 kbp ) and downstream ( +20/+75 kbp ) of their TSSs ( Figure 1F ) , when compared to housekeeping ( CE ) genes . No significant difference in the distribution of acetylated SINEs was observed in NR genes , when compared to CE genes ( Figure 1F ) . Due to the relatively low number of NR genes ( 70 ) the profile of acetylated SINEs is considerably scattered and does not allow a robust statistical analysis . The analysis of the genomic regions surrounding the TSSs of NEE-induced genes revealed that although acetylated SINEs were significantly enriched at a distance greater than 20 kbp from the TSS ( Figure 1F ) , in many cases , they were also found between −20 and +20 kbp from the TSS ( Table 1 ) . It should be noted that additional acetylated SINEs were discovered between 20 and 100 kbp from the TSS for all genes listed in Table 1 ( Figure 1F and LC and AR , unpublished results ) . Our findings indicate that de novo acetylation of SINEs represents a landmark of inducible genes , suggesting that these elements may play a critical role in regulating activity-dependent transcription . In neurons , regulatory sequences with enhancer functions are found within a wide range of distances from the TSS . The c-Fos gene for example , contains at least five enhancers located between −40 and +15 kbp from the TSS [16] . Neuronal enhancers are characterized by both the presence of specific epigenetic modifications , including H3K4Me1 , and by the binding to specific nuclear factors , such as CREB , SRF , Npas4 and CBP . To investigate whether acetylated SINEs may represent a new class of regulatory elements that control synaptic activity-dependent transcription in neurons , we analysed c-Fos and Gadd45b , two genes that fulfilled several important criteria: First , they are both activity-dependent genes with well-known functions in the nervous system [31] , [43]; Second , following 45 minutes of exposure to NEE , both genes undergo rapid and robust transcriptional activation in the cortex ( 6 . 0±0 . 8 and 2 . 7±0 . 7 fold induction , respectively , as assessed by qRT-PCR ( Figure S1C ) ) ; Third , ChIPseq analysis of the genomic regions surrounding the c-Fos and Gadd45b TSSs using somatosensory cortices of mice maintained in resting conditions or exposed to NEE revealed a stable acetylation profile , with the relevant exception of two regions encompassing RSINE1 and B1F , two SINEs located 3 . 9 and 9 . 2 kbp downstream from the c-Fos and Gadd45b TSSs , respectively ( Figure 2A , B and S1H ) . It should be noted that additional de novo acetylated SINEs are present within 100 kbp from the TSS of both c-Fos and Gadd45b ( Figure 1F ) . ChIP experiments confirmed that in somatosensory cortex , both c-FosRSINE1 and Gadd45bB1F underwent de novo H3K9K14 acetylation in response to NEE , while the acetylation levels of promoter regions remained unchanged ( Figure 2C ) . Similarly , c-FosRSINE1 and Gadd45bB1F became acetylated when cortical neurons were exposed to potassium chloride ( 50 mM , 45 minutes ) , a paradigm of stimulation that induces neuronal depolarization , calcium influx and gene transcription , that is commonly used to mimic synaptic activation in vitro ( Figure S2 ) [16] , [44]–[46] . Although the de novo acetylated SINEs did not present DNA elements that may mediate the binding of nuclear factors previously identified as interacting with enhancers [16] , the SINEs B box represents the consensus sequence for the multisubunit general transcription factor TFIIIC ( Figure S1H ) [24] . TFIIIC is part of the RNAPIII complex , and is best characterised in the context of its role in mediating the recruitment of the transcriptional complex to the B box located within the promoters of RNAPIII- transcribed genes [24] . TFIIIC comprises six subunits , some of which have known functions . For example , Gtf3c1 mediates the binding to the B box [24] , while Gtf3c2 and Gtf3c4 possess histone acetyltransferase activity ( HAT ) in vitro [47] . Importantly , TFIIIC may interact and recruit the HAT p300 to tRNA gene promoters [48] . Recent ChIP-chip analyses performed in fission yeast demonstrated that TFIIIC subunits are detected in the absence of RNAPIII at many genomic sequences known as Extra-TFIIIC sequences ( ETCs ) [49] , [50] . In yeast , binding of TFIIIC to ETC sites is necessary for tethering distant loci to the nuclear periphery , thereby contributing to the maintenance of the three-dimensional structure of chromosomes [49] , [51] . ChIPseq screens performed in both human and mouse cells also identified a large number of ETC sites that are frequently located proximal to RNAPII-dependent genes [25]–[27] . Moreover , in HeLa cells , at least 181 SINEs contain ETC sites [26] . We first performed an in silico analysis to verify whether +Δac regions contained putative ETCs . A consensus sequence for a “B box-like ETC” resembling the B box located at RNAPIII promoters , was recently identified with a ChIPseq screen targeting a subunit of TFIIIC [25] . We employed NestedMICA to determine whether the B box-like ETC motif was over-represented within de novo acetylated regions . +Δac regions showed a significant over-representation of the B box-like ETCs when compared either to a random set of genomic sequences ( p = 0 . 0029 ) or to a random set of sequences overlapping gene TSSs ( p = 0 . 0032 ) . Because +Δac regions harbour both putative TFIIIC binding sites and acetylated SINEs , we asked whether TFIIIC complexes interacted with de novo acetylated SINEs located in the proximity of activity-dependent genes . Indeed , the position-weight matrix of the B box-like ETC [25] predicted a putative TFIIIC binding site at the c-FosRSINE1 and Gadd45bB1F elements located downstream of c-Fos and Gadd45b TSSs , respectively ( Figure 2A , B , Figure S1H and Table 1 ) . Further analysis of acetylated SINEs located within 20 kbp from the TSS of NI genes consistently identified putative binding sites for TFIIIC ( Table 1 ) . ChIP experiments on mouse somatosensory cortex and primary cortical neurons confirmed that the TFIIIC subunit Gtf3c1 binds to both c-FosRSINE1 and Gadd45bB1F at levels comparable to the 5s rRNA gene ( Figure 2D , E ) . Conversely , Gtf3c1 binding was not detected at Jdp2B1 or GapdhB4 , two SINEs located 9 and 15 kbp from the TSS of a NR and a CE gene respectively , that lack a conserved B box and are located in genomic regions not associated with H3K9K14ac tags ( Figure 2D , E; Genomic coordinates of SINEs analysed are listed in Table S3 ) . To study whether dynamic changes of TFIIIC binding may represent a common feature of activity-dependent regulatory elements , neurons were exposed to KCl or left untreated and Gtf3c1 binding was tested on Gadd45bB1F and c-FosRSINE1 . As a control , five previously identified c-Fos enhancers were analysed [16] . We found that TFIIIC interaction with Gadd45bB1F and c-FosRSINE1 increased in response to depolarization . In contrast no significant binding of Gtf3c1 was observed on c-Fos enhancers ( Figure 2F ) . Interestingly , both p300 and the neuronal enhancer-associated histone marker H3K4Me1 were dynamically regulated on c-FosRSINE1 in response to depolarization ( Figure 2G , CP and AR , unpublished observations ) . Although CBP is recruited to c-Fos enhancers [16] , it was not detected on Gadd45bB1F and c-FosRSINE1 . These findings , together with the observation that Gadd45bB1F and c-FosRSINE1 are transcribed in response to depolarization ( CP and AR , unpublished observations ) suggest that acetylated SINEs may represent a novel class of regulatory elements that control the expression of activity-dependent genes in neurons . The finding that TFIIIC interacted with SINEs that became acetylated in response to depolarization prompted us to investigate whether inhibition of TFIIIC influenced activity-dependent transcription . Mouse cortical neurons were transfected with either Gtf3c5 or control siRNA , depolarised with KCl and c-Fos mRNA levels were measured by quantitative RNA-FISH . The yeast homolog of Gtf3c5 stabilises the interaction of TFIIIC with the B box [52] , although it does not directly bind to DNA . Transfected cells were identified by co-expression of a GFP vector . In non-stimulated neurons transfected with control siRNA , c-Fos transcripts were not detectable . In response to depolarization , discrete c-Fos mRNA ribonucleoparticles appeared in proximity to the nucleus ( Figure 3A , upper panels and Figure S3A , B ) and were clearly detectable in approximately 60% of stimulated neurons . This finding is in agreement with a recent report showing that depolarization of rat cortical neurons induced Arc expression only in about 50% of cells [14] . Similarly , in Dictyostelium discoideum , single cell analysis of IEGs transcription showed a high degree of variability between cells [53] . Quantitative fluorescence intensity analysis of c-Fos-containing ribonucleoparticles demonstrated that both in non-stimulated neurons and in response to depolarization , silencing of Gtf3c5 increased c-Fos transcription , when compared to control siRNA ( Figure 3A , B ) . Interestingly , Gtf3c5 silencing increased fluorescence intensity rather than the number of cells containing detectable c-Fos signal ( LC and AR , unpublished observations ) . In neurons transfected with Gtf3c5 siRNA , stimulus-independent increase of c-Fos expression was observed even in the presence of nifedipine , demonstrating that it was not due to depolarization events that may have been triggered by the silencing of Gtf3c5 ( Figure S3C ) . To investigate whether TFIIIC inhibition increased the expression of activity-dependent genes via transcriptional or posttranscriptional mechanisms , neurons were infected with lentiviral vectors driving the simultaneous expression of short hairpin RNA targeting either Gtf3c5 or firefly luciferase and GFP . Three days after infection , neurons were stimulated with KCl and subjected to qRT-PCR analysis . In neurons exposed to KCl , inhibition of Gtf3c5 was sufficient to enhance activity-dependent transcription of both pre-mRNA and fully processed mRNA of c-Fos ( Figure 3C , D ) and Gadd54b ( Figure 3E , F ) . In contrast , the expression of 5s rRNA , a TFIIIC- and RNAPIII-dependent transcript , was not affected ( Figure S3D ) . Although in neurons transfected with Gtf3c5 siRNA significant levels of c-Fos transcript were observed by RNA FISH in the absence of stimulation ( Figure 3A , B ) , a similar increase was not detected by qRT-PCR , when neurons were infected with shGtf3c5 . This discrepancy was probably due to the relatively low infection efficiency achieved in this set of experiments , which resulted in incomplete silencing of Gtf3c5 ( Figure S3D ) . In contrast , single-cell RNA-FISH exclusively allows the analysis of transfected neurons , greatly improving the accuracy of mRNA quantification . In agreement with the observation that TFIIIC inhibition disrupts the regulated transcription of activity-dependent genes , c-FosRSINE1 and Gadd45bB1F acetylation was higher in neurons infected with shGtf3c5 under both basal and stimulated conditions ( Figure 3G ) . The increase of SINE acetylation was even more remarkable when infection efficiency of shGtf3c5 lentivirus and Gtf3c5 inhibition ( both around 60% ) were taken into account . Acetylation of c-Fos promoter and GapdhB4 was unchanged , indicating that inhibition of TFIIIC specifically affects regulatory SINEs ( Figure 3G ) . It should be noted that expression levels of TFIIIC subunits were not changed in the cortices of mice exposed to NEE or in cortical neurons stimulated with KCl ( Figure S3E , F ) . Moreover , KCl depolarization did not affect neuronal viability ( Figure S3G ) . Taken together , these findings demonstrate that inhibition of TFIIIC induces unregulated transcription and enhances gene expression both in resting condition and in response to depolarization . Thus , TFIIIC may act as a “brake” of transcription that contributes to maintaining activity-regulated genes inhibited in the absence of stimulation . In yeast , binding of TFIIIC to ETCs regulates the three-dimensional arrangement of chromatin , resulting in tethering of chromosomal regions to discrete clusters at the nuclear periphery , where chromatin is maintained in a transcriptional repressed state [49] , [51] . The eukaryote genome harbours several hundred ETCs of unknown function [25]–[27] that similarly to yeast may play a role in positioning genes within the nucleus , thereby regulating their expression . Visualization of transcription sites in mammalian nuclei has revealed that RNAPII complexes are distributed in discrete foci known as transcription factories ( TFs ) [19] , [54] . Importantly , in mouse B lymphocytes , the relocation of genes to TFs mediates stimulus-dependent transcription [55] . We first studied whether relocation of activity-regulated neuronal genes to TFs may be detected in nuclei of cortical neurons using immuno-DNA FISH . This technique allows the simultaneous detection of gene loci and RNAPII foci without affecting the three-dimensional structure of the nucleus ( Figure 4A ) . Immunostaining of mouse cortical neurons with an antibody that recognizes a form of RNAPII associated with transcriptional initiation ( RNAPII phospho-serine 5 ) [17] , [54] detected 217±50 TFs ( n = 27 cells ) . This finding is in agreement with the number of TFs observed in mouse fetal brain [54] . Computational analysis of the confocal image stacks was employed to quantify the degree of co-localization between gene loci and TFs . The distance between the centre of each DNA-FISH signal and the nearest TF was measured , and the threshold for co-localization was set at 225 nm . This was the distance at which two of the smallest detectable objects were considered overlapping . In the nuclei of cortical neurons exposed to KCl , co-localization of both c-Fos and Gadd45b loci with TFs increased dramatically ( Figure 4A–C ) . Similar results were observed when DNA FISH experiments were performed using cortical neurons cultured for 10 days and exposed to bicuculline , an antagonist of GABAA receptors that enhances glutamatergic transmission ( Figure 4D and S4A ) . Bicuculline- and KCl-dependent relocation to TFs was completely abolished by blocking depolarization with tetrodotoxin ( TTX ) or nifedipine , respectively ( Figure 4D , E and S4A , B ) . Importantly , relocation of c-Fos was also observed when neurons were treated with the transcription inhibitor 5 , 6-dichloro-1-beta-D-ribofuranosylbenzimidazole ( DRB; Figure 4E and S4B ) , confirming that TFs are stable subnuclear structures that exist independently of active transcription both in fixed and in living cells [56] , [57] . Consistent with the hypothesis that relocation to TFs is a necessary step for the expression of most , if not all , activity-regulated genes , and in agreement with a recent study [58] , we found that Bdnf , an activity-dependent gene that was induced upon NEE , relocated to TFs in response to depolarization . Although the Bdnf gene does not have an acetylated SINE located within 100 kbp from its TSS , it is possible that SINEs located further from the promoter may result in looping of the chromatin that induces relocation of Bdnf to TFs . Indeed , several studies have shown that chromatin loops promoted by distal regulatory elements , situated several hundreds kbp away from the TSSs of their target genes , are capable of triggering the relocation of gene loci to active chromatin hubs [59]–[61] . Similarly , an acetylated SINE was detected 2 kbp upstream the TSS of Fcf1 , a gene that did not relocate to TFs following depolarization and was not expressed in response to neuronal activity ( Figure 4A–C ) . Thus , as for most regulatory elements , the distance of acetylated SINEs from TSS is not necessarily predictive of their regulatory function . As additional controls , we tested the loci of both Csn2 , a gene that is not expressed in primary cortical neurons and Gapdh , a gene that is costitutively expressed , and found that they did not relocate to TFs in response to depolarization ( Figure 4A–C ) . Thus , in neurons , as for other cell types [55] , relocation of gene loci to TFs represents a necessary event that mediates activity-dependent transcriptional activation . To study whether TFIIIC regulates the tethering of inducible gene loci to TFs , the Gtf3c5 subunit of TFIIIC was silenced by siRNA transfection ( Figure S5A ) . Transfected neurons were identified by co-transfection of an expression vector encoding eGFP-tagged actin that remains confined in the cytoplasm , allowing the detection of fluorescein-labeled DNA-FISH probe within the nucleus ( Figure 5A ) . In resting conditions , transfection of Gtf3c5 siRNA increased the co-localization of c-Fos and Gadd45b loci with TFs , at levels comparable to cells transfected with control siRNA and exposed to KCl ( Figure 5A , B and S5B , C ) . Gtf3c5 silencing did not affect the co-localization of the silent gene Csn2 ( Figure 5B and S5D ) . Interestingly , in neurons transfected with Gtf3c5 siRNA , KCl treatment did not further increase relocation of c-Fos and Gadd45b loci to TFs ( Figure 5B and S5B , C ) , indicating that knockdown of a TFIIIC subunit entirely mimicked the effects of neuronal depolarization on transcription . Enhanced relocation of c-Fos under basal conditions was also observed in neurons cultured in the presence of nifedipine , demonstrating that it was not dependent on depolarization events that may have been triggered by the silencing of Gtf3c5 ( Figure S5E ) . The specificity of Gtf3c5 silencing was confirmed by generating a “rescue” siRNA-resistant form of Gtf3c5 ( myc-Gtf3c5R; Figure S5F , G ) . Cortical neurons were transfected with expression vectors encoding myc-Gtf3c5R with shRNA targeting either Gtf3c5 ( shGtf3c5 ) or firefly luciferase ( shLuc , as a negative control ) and subjected to immuno-DNA FISH . In resting neurons , expression of myc-Gtf3c5R completely rescued the effect of Gtf3c5 silencing on c-Fos relocation to TFs ( Figure S5H ) . Our findings indicate that in the absence of stimulation , TFIIIC inhibits c-Fos and Gadd45b transcription by preventing their relocation to TFs . In response to depolarization , genes undergo nuclear repositioning to TFs and this event creates a molecular context highly permissive to transcription . Neurons maintained in chronic depolarizing conditions undergo a dramatic increase of both dendritic length and branching [44] , [45] and prolonged exposure to NEE ultimately leads to neuronal differentiation and dendritogenesis [1] . To investigate whether TFIIIC-dependent transcription influenced dendritic growth , mouse cortical neurons were transfected with an expression vector encoding GFP alone or in combination with control or Gtf3c5 siRNA . Neurons were left untreated or exposed to depolarizing conditions ( 50 mM KCl ) for two days . Sholl analysis showed that in neurons exposed to KCl dendrites were more complex ( Figure 6A , B and Figure S6 ) and total dendritic length was increased ( Figure 6C ) . In contrast , neurons transfected with Gtf3c5 siRNA showed a dramatic increase of dendritogenesis when maintained in resting , non-depolarizing conditions . Remarkably , in neurons maintained in basal conditions , dendritic arborization and total dendritic length were comparable to neurons subjected to chronic KCl treatment that were either non-transfected or transfected with control siRNA ( Figure 6 ) . In neurons transfected with Gtf3c5 siRNA , depolarization also significantly enhanced dendritogenesis ( Figure S6D ) . Interestingly , nifedipine failed to completely inhibit depolarization-dependent transcription of c-Fos in neurons transfected with Gtf3c5 siRNA ( Figure S3C ) . Because Gtf3c5 silencing induced unregulated transcription of many genes that control neuronal functions , it is possible that as a result , these neurons presents both stronger synaptic contacts and more robust glutamatergic signaling , making them less sensitive to nifedipine . The growth and development of dendrites depends on de novo transcription of a number of genes , which are usually inactive in resting conditions [44] . Impairment of TFIIIC in non-stimulated neurons was sufficient to induce extensive dendritic growth comparable to chronic depolarization , indicating that upon Gtf3c5 silencing , many genes normally expressed only in response to depolarization underwent unregulated transcription .
Under physiological conditions the ever-changing environment constantly challenges the brain , which must respond with rapid adaptive strategies to ensure survival . For this to occur , it is essential that hundreds of activity-dependent genes become activated in an accurate and timely manner . Since the identification of c-Fos as the prototypical neuronal activity-dependent gene [31] , the molecular mechanisms of transcriptional responses to depolarization have been extensively studied [8] , [9] , [62] . Following synaptic activity , increase of intracellular calcium levels triggers the initiation of signaling pathways that lead to the activation of transcription factors and their binding to specific gene promoters [30] , [62] , [63] . This event , together with the recruitment of histone modifying enzymes to chromatin , results in the rapid transcription of target genes [8] , [9] , [62] . In addition to promoters , many other regulatory sequences are scattered throughout the genome , with both transcriptional activating ( enhancers ) and repressing ( silencers and insulators ) functions [64] . In neurons , a novel class of enhancers associated with the p300 homolog CBP was recently shown to regulate activity-dependent gene expression [16] . Several models have been proposed to explain the molecular mechanisms by which enhancers activate gene expression , including chromatin looping that may favour both promoter-enhancer interaction and their relocation to transcriptionally active subnuclear compartments . We discovered that SINEs located in the proximity of inducible gene promoters represent a new class of regulatory elements that coordinate activity-dependent transcription . A genome wide ChIPseq screen performed using somatosensory cortex of mice exposed to NEE demonstrated that promoters of rapidly activated genes , including c-Fos and Gadd45b are remarkably stable and do not undergo changes of H3K9K14ac in response to synaptic activity . In contrast , SINEs located between −100 and +100 kbp from the TSS of inducible gene promoters became rapidly acetylated . Previous studies have demonstrated that SINEs , in addition to their established role as insulators [65] , [66] , may also influence gene expression in the brain . For example , AmnSINEs , a highly conserved family of SINEs identified in the genome of amniota , contribute to the expression of genes necessary for brain development , in some cases acting as tissue-specific enhancers [39] , [40] . Although acetylated SINEs shared some properties with previously identified neuronal enhancers [16] , including the presence of H3K4me1 and increased transcription in response to depolarization , they do not include the consensus sequences for the nuclear factors CREB , SRF or Npas4 ( CP and AR , unpublished observations ) . Acetylated SINEs contained a motif that completely overlapped with the B box , the binding sequence for the general transcription factor TFIIIC [50] . Importantly , TFIIIC binding to acetylated SINEs located near the TSS of inducible genes regulated activity-dependent transcription . Silencing of the GTf3c5 subunit induced both SINE acetylation and unregulated transcription of inducible genes that resulted in a dramatic increase of dendritic growth and branching . Thus , the interaction of TFIIIC with acetylated SINEs may act as a “brake” of the transcriptional response , coordinating the expression of the many genes required for dendritic growth and neuronal differentiation . How does TFIIIC control gene relocation and transcription in neurons ? In mammalian cells , genes that are activated either in response to stimulation or during cell differentiation preferentially relocate to shared TFs [22] , [55] . The simultaneous relocation of activated genes to preassembled TFs enriched with RNAPII and other specific nuclear factors [22] ensures accurate and efficient transcription [19] . We found that the interaction of TFIIIC with SINEs located near activity-dependent neuronal genes controlled their relocation to TFs , where they were rapidly transcribed . The enhanced binding of Gtf3c1 and p300 at SINEs observed in response to depolarization ( Figure 2F , G ) may reflect the simultaneous relocation and clustering of several SINE and promoter elements at shared TFs , where the local concentration of transcription factors and co-activators increases significantly ( Figure 7A ) . Two subunits of the TFIIIC complex possess acetyltransferase activity [47] and we and others have found that the histone acetyltransferase p300 is recruited to TFIIIC binding sites ( Figure 2G and [48] ) . Therefore , TFIIIC may mediate de novo acetylation of SINEs either directly or indirectly via interaction with p300 . Although in non-stimulated neurons TFIIIC functions as a transcriptional “brake” , a subunit switch may occur within the complex in response to neuronal activity , which allows TFIIIC to become a platform for the recruitment of transcriptional co-activators ( Figure 7B ) . As suggested by the silencing experiments , the dissociation of the Gtf3c5 subunit from the complex may induce such a functional switch . An additional mechanism may rely on the ability of TFIIIC to mediate the loading of condensins on DNA [67] . In all eukaryotes , condensins are tightly associated with chromatin and regulate the assembly and the maintenance of chromosomal structure during mitosis [68] , [69] . Interaction of condensins with TFIIIC bound to SINEs , located near activity-dependent neuronal genes , may mediate rapid changes of nuclear structure that results in the release of gene loci from the nuclear periphery , where they are maintained in a repressed state , to TFs , where they are transcribed . Thus , our data provide a link between TFIIIC function and transcriptional activation in response to depolarization , and identify a new role for the hundreds of ETC sites scattered across the mammalian genome [25] , [27] . Finally , in the nervous system , nuclear architecture undergo substantial modifications during the different stages of differentiation , presumably in response to developmental cues [70] . Interestingly , three-dimensional image reconstruction analyses demonstrated that nuclei of hippocampal neurons undergo dramatic infolding and possibly substantial chromatin re-organization following short-term burst of synaptic activation [71] and in humans , the X chromosome shows a dramatic nuclear repositioning in response to epileptic seizures [72] . One of the implications of our study is that in neurons , changes of nuclear architecture are required not only for long-lasting expression of specific genes in differentiated cells , but also for more rapid transcriptional responses .
C57BL/6J female mice were separated after weaning and housed singly in conventional shoebox polycarbonate cages with ad libitum access to food pellets and water . Female mice ( 2 to 4 months of age ) were used in order to avoid any aggressive behaviour caused by mating or territorial needs . Prior to the experiments , mice were gently handled for a few minutes every day for at least a week to minimize the stress associated with handling and to ensure that the mice were exclusively stimulated by exposure to enriched environmental conditions . On the day of the experiment , mice were randomly sorted into control ( e . g . unstimulated ) or stimulated groups , and stimulation was performed always in the late afternoon . Stimulation was achieved by moving the mice to a complex environment represented by two large , semi-transparent plastic boxes ( 80×40×50 cm each ) joined by three pipes . The boxes were fitted with cardboard tubes , mazes and several structures to provide spatial complexity . Several items ( including wood chips and paper towels ) and food ( seeds , dried and fresh fruit ) were used for sensory stimulation . Social interactions were obtained by stimulating at least five mice simultaneously . While in the novel environment , mice displayed a consistent exploratory behaviour , occasionally feeding and drinking , and were constantly monitored for signs of stress ( e . g . fighting , stereotypical behaviours ) . Mice were exposed to NEE for the indicated times , culled by cervical dislocation and brain dissection was performed immediately after culling . Dissected cortices were immediately frozen into dry ice-cooled isopentane and stored at −80°C . Control animals were culled on the day of the stimulation . Cortices of E15 mouse embryos were dissected in ice-cold HBSS , incubated with 20 U/ml Papain ( Worthington ) for 25 minutes at 37°C . Tissue was resuspended in adhesion medium ( MEM supplemented with 10% FBS and 5% horse serum ) and gently dissociated with a 5 ml serological pipette . Neurons were plated either on glass coverslips in 4 well-plates or on 10 cm dishes , and after 4 hours , culture medium was replaced with Neurobasal medium supplemented with B27 . For Immuno-DNA FISH and RNA FISH experiments , neurons cultured on glass coverslips for 2 days were transfected using Optimem containing 200 ng of plasmidic DNA , 100 pmol of siRNA and 0 . 8 µl Lipofectamine2000 ( Invitrogen ) . Either control siRNA ( QIAGEN Allstar negative control , #1027281 ) or a pool of Gtf3c5 siRNA molecules ( ON-TARGETplus SMARTpool L-056126-01-0005 , Dharmacon; target sequences: gag gaa agc guc ucu cga a , gca cca auc cca uag auc a , acu gau ggc cca cgg aaa u , aca gag ugc uca ugc gca a ) were used , in combination with an eGFP-actin or GFP expression vector . After 3 hours , transfection medium was replaced with culture medium and neurons were cultured for 72 hours before stimulation . Cells were either stimulated with 50 mM KCl for 45 minutes or left untreated . Prior to stimulation , cells were cultured overnight in low serum ( 30% compared to standard medium ) in the presence of 50 µM AP5 . Unless otherwise stated , Nifedipine ( MP Biomedicals ) , TTX ( Tocris Bioscience ) and DRB ( Sigma ) were added to the culture medium 15 minutes prior to stimulation . Bicuculline was purchased from Tocris Bioscience . NIH/3T3 fibroblasts were cultured according to to ATCC guidelines and transfected using Lipofectamine2000 ( Invitrogen ) as per manufacturer's instructions . Cell culture reagents were purchased from Gibco . shRNA sequences targeting either firefly Luciferase ( shLuc , gct gac gcg gaa tac ttc gtt caa gag acg aag tat tcc gcg tca gc ) or Gtf3c5 ( shGtf3c5 , gag gaa agc gtc tct cga att caa gag att cga gag acg ctt tcc tc ) were cloned into pSUPER . neo+gfp vector ( BglII/XhoI ) . To obtain a myc-tagged isoform , mouse Gtf3c5 coding sequence was cloned from pCMV-SPORT6-Gtf3c5 ( Image clone IRAVp968E0646D , Source Bioscience ) into pCMV-myc by blunt-end cloning and subsequently the siRNA-target region was mutagenized ( referred to as myc-Gtf3c5R; Quickchange II site-directed mutagenesis , Stratagene ) . GFP coding sequence in pSUPER constructs was either excised ( pSUPER-shLuc and pSUPER-shGtf3c5 constructs ) or substituted with myc-Gtf3c5R coding sequence ( Nhe/NotI ) , allowing the simultaneous silencing of endogenous Gtf3c5 and the expression of shRNA-resistant myc-Gtf3c5R . shLuc and shGtf3c5 sequences were cloned into the GFP-containing lentiviral vector L303 ( XhoI/XbaI; gift of A . Citri ) . Self-inactivating HIV lentivirus particles were produced by transfecting HEK293T cells with the vector , envelope and packaging plasmids ( pCMV_VSV_G and Δ8 . 9 , Addgene ) . Mouse primary cortical neurons were infected 6 hours after plating , and medium was changed after 16 hours . 5 days after infection , neurons were either stimulated with 50 mM KCl for 45 minutes or left untreated , and subjected to RNA extraction , cDNA synthesis and expression analyses . Somatosensory cortices dissected form either control or NEE-exposed mice were fixed in 1% formaldehyde in PBS for 15 minutes . Cross-linking reaction was terminated by adding glycine to a final concentration of 125 mM . After homogenisation , dissociated cells were harvested by centrifugation and resuspendend in sonication buffer ( 20 mM Tris-HCl pH 8 . 1 , 150 mM NaCl , 0 . 1% SDS , 0 . 5% Triton X-100 ) . Alternatively , dissociated cells were resuspended in cell lysis buffer ( 5 mM PIPES pH 8 . 0 , 85 mM KCl , 0 . 5% NP40 ) and incubated up to 15 minutes on ice . Nuclei were purified by centrifugation ( 5 minutes at 1000×g ) , lysed in nuclei lysis buffer ( 50 mM Tris-HCl pH 8 . 1 , 10 mM EDTA , 1% SDS ) , and incubated 20 minutes on ice . Lysates were frozen in liquid N2 and thawed twice prior to sonication . For in vitro experiments , at least 5×10∧6 primary cortical neurons were fixed in 1% formaldehyde in PBS for 20 minutes , followed by inactivation with 125 mM glycine . Cells were harvested in 100 mM Tris-HCl pH 9 . 4 , 10 mM DTT , centrifuged and resuspended in sonication buffer . Sonicaton was performed using a Bioruptor ( Diagenode ) , by applying 30 pulses , 30 seconds each , at 30 seconds intervals , thus shearing chromatin into fragments 200–400 nt in size . After centrifugation at 13 , 000 g for 10 minutes at 4°C , supernatants were pre-cleared with Protein A-Sepharose beads ( GE Healthcare , 1 hour at 4°C ) prior to immunoprecipitation with anti-Histone H3K9K14ac ( 06-599 , Millipore ) , anti-Histone H3 ( ab10799 , Abcam ) , anti-Gtf3c1 ( A301-291A , Bethyl Laboratories ) , anti-p300 ( Santa Cruz , sc-585X ) or control IgG ( Santa Cruz ) , overnight at 4°C . Immune complexes were purified by incubating the lysates with Protein A-Sepharose beads for 1 hour at 4°C , followed by 5 washes in 50 mM Hepes pH 7 . 6 , 1 mM EDTA , 0 . 5 M LiCl , 0 . 7% sodium deoxycholate , 1% NP40 , and 2 washes in 10 mM Tris-Hcl ph 8 . 1 , 1 mM EDTA , 5 minutes each . All buffers contained protease and phosphatase 2 and 3 inhibitor cocktails ( Sigma ) . Beads were boiled with 10% Chelex 100 resin ( Bio-Rad ) and DNA was eluted with 10 mM Tris-HCl pH 8 . 0 . Alternatively , DNA was eluted by washing the beads twice for 15 min in 0 . 1 M NaHCO3 ( pH 8 . 0 ) , 1% SDS , followed by 16 hours incubation at 65°C in the presence of 300 mM NaCl and subsequent purification by PCR purification columns ( Qiagen ) , according to manufacturer's instructions . Total input DNA was purified by boiling for 15 minutes an aliquot of sonicated chromatin in the presence of 300 mM NaCl , followed by phenol-chlorophorm extraction . DNA was subjected to quantitative real-time PCR in 25 µl reactions containing 12 . 5 µl of DyNAmo Flash SYBR Green qPCR Kit ( Thermo Scientific ) and 0 . 2 µM primers . All reactions were performed in triplicate with an Eppendorf Mastercycler Realplex 2 and each experiment included a standard curve and a no-template control . Standard curves consisted of serial dilutions of gel-purified PCR amplicons of known concentration . At the end of 40 cycles of amplification , a dissociation curve was obtained in which SYBR Green fluorescence was measured at 1°C intervals between 60°C and 100°C . Melting temperatures of amplicons varied between 79 and 83°C . For Gtf3c1 ChIP experiments ( Figure 2D , E ) , negative control primers were designed on a 2 kbp region of chromosome 12 devoid of any H3K9K14ac and SINE . Primer sequences are listed in Table S4 . Chromatin was purified from the somatosensory cortex of mice either unstimulated or exposed to NEE for 45 minutes . Steps were taken to reduce potential sources of variability due to individual animals and antibody efficiency . Four ( control ) or six ( NEE ) age-matched mice were used for each experimental condition . Rather than performing a single immunoprecipitation for each sample , aliquots of chromatin were pooled into 4 separate immunoprecipitation reactions taking care to use the same amount of chromatin from each cortex . The genomic DNA obtained from the 4 immunoprecitations was pooled and subjected to high-throughput sequencing . For each condition , at least 7 millions tags were sequenced using the Illumina GA-II platform at Fasteris SA ( Switzerland ) . Reads were mapped to the mouse genome ( release NCBIm37 ) using MAQ 0 . 6 . 6 ( http://maq . sf . net/ ) . Reads with potentially ambiguous mappings were discarded ( MAQ quality score threshold > = 10 ) . To avoid bias due to potential clonality artifacts , we also discarded any second and subsequent mappings to a given position on the genome ( i . e . all mappings used in subsequent analyses must be derived from independent molecules ) . To determine regions of differential acetylation ( referred here as +Δac and −Δac ) all reads mapping within 2 kb sliding windows across the genome were counted and differential enrichment relative to a Binomial null model was analyzed ( parameters used: p = total_size_of_library_1/[total_size_of_library_1 + total_size_of_library_2] ) . This approach excluded the detection of artifacts deriving from the presence of non-specific immunoprecipitated DNA , thus making a negative control ChIP unnecessary . The number of reads ( Table S1 ) was sufficient to provide statistical robustness , as the false discovery rate estimated from the binomial model employed to identify the regions of differential acetylation was 0 . 040 . Windows with a score of p<10∧-3 were considered for subsequent analysis . Total RNA from somatosensory cortices of mice or from cultured primary cortical neurons were extracted using Trizol ( Invitrogen ) , according to the company specifications . Genomic DNA was removed by treating the samples with DNAse I ( Roche ) for 30 minutes , followed by phenol/chloroform purification . For expression analysis by quantitative real-time PCR , 1–2 µg of total RNA were reversed-transcribed by using Superscript III ( Invitrogen ) and analysed by using DyNAmo Flash SYBR Green qPCR Kit ( Thermo Scientific ) , as performed for ChiP-qPCR experiments ( see above ) . Primer sequences are listed in Table S4 . For the microarray experiments , age-matched female mice were separated into control and NEE-stimulated groups , and 3 mice were used for each experimental condition . Expression analysis was performed by Cambridge Genomic Services ( Cambridge , UK ) . Quality of total RNA was verified using Agilent 2100 Bioanalyzer followed by expression analysis using a MouseWG-6 Expression BeadChip ( Illumina ) . To identify DNA motifs over-represented in +Δac regions the motif prediction tool NestedMICA was employed [35] . A background model was generated using randomly selected genomic sequences of size comparable to +Δac regions . For sets of +Δac regions overlapping TSSs , the background sequences were also encompassing TSSs . The pool of analysed regions was randomly split into “test” and “validation” sets . Motif prediction was performed on the test set . To exclude false positives , over-representation of the obtained motifs was verified in the validation set . Predicted motifs showing over-representation in both test and validation sets , as compared to the background , were considered for further analysis . Immuno-DNA FISH experiments were performed as previously described [73] with modifications . Briefly , cells were fixed for 30 minutes in PBS containing 0 . 5% glutaraldehyde , 0 . 3% Triton-X 100 , followed by two 15 minutes treatments with 7 . 5% NaBH4 . After blocking with 5% normal goat serum and 5% fetal bovine serum , RNAPII-ser5P ( 4H8 , 05-623 Millipore ) and GFP ( ab6556 , Abcam ) antibodies were applied , followed by detection with secondary antibodies detection ( Anti-mouse Alexa568 , anti-rabbit Alexa488 , Molecular Probes ) , post-fixation with 50 mM EGS for 30 minutes at 37°C and 100 mg/ml RNAse A treatment for 1 hour . Between each step , coverslips were washed extensively with PBS . Nuclear genomic DNA denaturation was obtained by incubating the coverslips in 70% formamide in 0 . 2× SSC for 30 minutes at 80°C , followed by hybridization with a digoxigenin-labeled probe for 16 hours at 38°C . Probes were labelled with digoxigenin-dUTP using a nick translation kit ( both from Roche ) . 0 . 1 µg DNA probe was pre-annealed with mouse Cot-1 DNA for 45 minutes at 37°C and denatured 10 minutes at 95°C immediately before hybridization . The following BAC clones ( CHORI ) were used to generate digoxigenin-labeled probes: c-Fos , RP24-233K8; Gadd45b , RP23-382P20; Csn2 , RP23-110B6; Gapdh , RP23-268O22; Bdnf , RP24-149F11; Fcf1 , RP24-185C13 . Confocal images of neuronal nuclei were acquired using a Leica TCS SP5 confocal microscope ( z-distance 0 . 2 µm ) . Images were analysed using Fiji and an appositely developed algorithm . No blinding was required as the analysis was performed computationally , with limited pre-processing of the images . The 3D Objects Counter tool was used to identify each DNA FISH signal and measure the coordinates of its centre of mass , based both on the signal shape and the intensity of each voxel . The 3D Objects Counter tool was employed also to identify transcription factories . RNAPII-ser5P foci were identified by applying a threshold based on mean and standard deviation of the voxels within the cell nucleus ( counterstained by DAPI ) , to account for the small variability of fluorescence intensity between different cells and experiments . Based on the threshold level ( calculated as mean +1 S . D . ) voxels were defined either as RNAPII “positive” or “negative” . Gene locus-to-TF distance was determined by using an algorithm that calculated the distance between the centre of mass of the DNA FISH signal and the nearest of the “RNAPII positive” voxels , in all directions and not only on the same confocal plane . The co-localization threshold was set to 225 nm , corresponding to the distance at which the two smallest detectable objects overlap . Co-localization of FISH signals and TFs was verified by visual inspection after the analysis . Mouse primary cortical neurons were fixed for 30 minutes with 4% PFA in PBS , followed by a 5 minutes wash with 0 . 1% DEPC in PBS and a 20 minutes permeabilization step with 0 . 3% Triton X-100 in PBS . After each step , coverslips were washed extensively with PBS . All solutions were RNAse-free . After pre-hybridization for 2 hours at room temperature in hybridization solution ( Sigma ) , coverslips were incubated for 16 hours at 55°C with hybridization solution containing 500 ng/ml of digoxigenin-labeled cRNA probe . Probes were labelled with digoxigenin-UTP by performing in vitro transcription ( Roche ) of a linearized plasmidic DNA template encoding c-Fos coding sequence ( NM_010234 ) . Prior to hybridization , probes were denatured for 5 minutes at 85°C . Excess probe was washed using increasingly stringent conditions ( 5 minutes at 55°C in 5× SSC , 1 minute at 55°C in 2× SSC , 30 minutes at 55°C in 0 . 2× SSC with 50% formamide , 5 minutes at room temperature in 0 . 2× SSC ) and coverslips were incubated with 1% Blocking reagent ( Roche ) in 250 mM NaCl , 100 mM Tris-HCl pH 7 . 5 for 1 hour , followed by consecutive incubation with primary ( anti-digoxigenin alkaline phosphatase conjugated from sheep , Roche; anti-GFP from rabbit , Abcam ) and secondary antibodies ( anti-rabbit Alexa488 , Invitrogen ) . Anti-digoxigenin immunoreactivity was detected by applying the alkaline phosphatase substrate Fast Red ( Roche ) , according to the manufacturer's instruction . Confocal images of neuronal soma were acquired using a Leica TCS SPE confocal microscope and analysed using Fiji . 3D Objects Counter tool was employed to identify c-Fos ribonucleoparticles and measure their fluorescence intensity . Ribonucleoparticles located within transfected cells only were identified using the GFP counterstaining . 3 hours after plating , mouse cortical neurons were transfected using Optimem containing 400 ng of GFP expression vector ( no siRNA control ) , or 200 ng GFP vector and 100 pmol of siRNA ( control and Gtf3c5 siRNA ) and 0 . 8 µl Lipofectamine2000 ( Invitrogen ) . After 3 hours , medium was replaced with culture medium with or without 50 mM KCl . Cells were cultured for 2 days followed by immunostaining with anti-GFP ( ab6556 , Abcam ) and anti-MAP2 ( M9942 , Sigma ) antibodies . Dendrites were identified by MAP2 staining . Images were obtained using a Zeiss Axioplan 2 microscope and analysed in Fiji . For Sholl analysis and quantification of total dendritic length we used the Simple Neurite Tracer plugin . Images of the dendritic profiles shown in Figure 6A were generated using the Trainable Segmentation plugin ( original images are provided in Figure S6A ) . Brains of control and NEE-exposed mice were frozen in dry ice-cold isopentane and stored at −80°C until cryosectioning . 12 µm coronal sections were cut using a Leica CM1850 cryostat . Sections were fixed in PBS containing 4% paraformaldehyde for 20 minutes , followed by 10 minutes incubation with 1% triethanolamine , 0 . 25% acetic anhydride and 2 h prehybridization with hybridization solution ( Sigma ) . Between each step , slides were washed extensively in PBS . Samples were incubated overnight at 55°C with 500 ng/ml of digoxigenin-labeled probe in hybridization solution and washed as for fluorescent in situ hybridization ( see above ) . Slides were incubated with 1% Blocking reagent ( Roche ) for 1 hour , followed by anti-digoxigenin-POD conjugate antibody and NBT/BCIP colorimetric reaction . Cortical neurons were cultured on glass coverslips , fixed with 4% PFA in PBS for 20 minutes , permeabilized with 0 . 3% Triton X-100 in PBS and subjected to immunostaining using anti Gtf3c5 ( A301-242A , Bethyl Laboratories ) , anti-GFP ( ab13970 , Abcam ) and anti-c-Myc ( ab32 , Abcam ) . Cell nuclei were identified using DAPI . Cells were harvested , resuspended in RIPA buffer ( 50 mM Tris pH 7 . 5 , 150 mM NaCl , 1 mM EDTA , 1% NP40 , 0 . 5% sodium deoxycholate , 0 . 1% SDS ) , incubated 20 minutes on ice and centrifuged 20 minutes at 13 , 000×g . Mouse cortex samples were homogenized in RIPA buffer and processed as above . Supernatants were denatured in Laemmli buffer followed by SDS-PAGE and western blotting analysis . Antibodies used were Gtf3c1 ( NB100-60657 , Novus Biologicals ) , Gtf3c2 ( A301-236A , Bethyl Laboratories ) , Gtf3c3 ( A301-238A , Bethyl Laboratories ) , Gtf3c4 ( A301-239A , Bethyl Laboratories ) , Gtf3c5 ( A301-242A , Bethyl Laboratories ) , Gtf3c6 ( ab107804 , Abcam ) , c-Myc ( sc56634 , Santa Cruz ) , Hsp90 ( sc1055 , Santa Cruz ) , Gapdh ( ab9494 , Abcam ) . All animal work must have been conducted according to relevant legislation in United Kingdom ( Animals Scientific Procedures Act 1986 ) . | In neurons , acetylation of histones and other epigenetic modifications influence gene expression in response to synaptic activity . Genes that are concomitantly expressed in response to stimulation are transcribed at specific nuclear foci , known as transcription factories ( TFs ) that are enriched with active RNA Polymerase II and often include specific transcription factors . Here , we show a novel regulatory role for Short Interspersed Elements ( SINEs ) located in the proximity of activity-regulated genes . SINEs represent a new class of regulatory sequences that function as coordinators of depolarization-dependent transcription . Binding of the general transcription factor TFIIIC to SINEs regulates activity-dependent transcription , relocation of inducible genes to transcription factories and dendritogenesis . Our study provides new fundamental insights into the mechanisms by which relocation of inducible genes to transcription factories and changes of nuclear architecture coordinate the transcriptional program in response to neuronal activity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cellular",
"structures",
"chromosome",
"structure",
"and",
"function",
"neuroscience",
"cell",
"differentiation",
"dna",
"transcription",
"histone",
"modification",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"epigenetics",
"cell",
"nucleus",
"c... | 2013 | Binding of TFIIIC to SINE Elements Controls the Relocation of Activity-Dependent Neuronal Genes to Transcription Factories |
Type 1 interferons ( T1-IFNs ) play a major role in antiviral defense , but when or how they protect during infections that spread through the lympho-hematogenous route is not known . Orthopoxviruses , including those that produce smallpox and mousepox , spread lympho-hematogenously . They also encode a decoy receptor for T1-IFN , the T1-IFN binding protein ( T1-IFNbp ) , which is essential for virulence . We demonstrate that during mousepox , T1-IFNs protect the liver locally rather than systemically , and that the T1-IFNbp attaches to uninfected cells surrounding infected foci in the liver and the spleen to impair their ability to receive T1-IFN signaling , thus facilitating virus spread . Remarkably , this process can be reversed and mousepox cured late in infection by treating with antibodies that block the biological function of the T1-IFNbp . Thus , our findings provide insights on how T1-IFNs function and are evaded during a viral infection in vivo , and unveil a novel mechanism for antibody-mediated antiviral therapy .
Type 1 interferons ( T1-IFNs ) are cytokines produced during viral infections by most infected cells and by some uninfected cells that recognize exogenous pathogen associated molecular patterns ( PAMPS ) through pattern-recognition receptors such as toll-like receptors . An important function of T1-IFNs is to stimulate the transcription of interferon stimulated genes ( ISGs ) through the nuclear translocation of the phosphorylated Stat1 transcription factor in cells which results in increased resistance to viral infection at the cellular and organismal level [1] . Many experimental approaches use systemic infection with viruses , which results in the rapid induction of systemic T1-IFNs and , consequently , ISGs . However , this type of infection rarely occurs in nature . To cause systemic disease , many viruses of importance to human and animal health such as viruses in the genera Orthopoxvirus ( OPV , variola ( VARV ) , monkeypox ( MPXV ) ) , Enterovirus ( polio , coxsackie ) , Aphtovirus ( foot-and-mouth disease ) , Rubivirus ( rubella ) , Flavivirus ( Yellow Fever , Dengue , West Nile ) , Rubulavirus ( mumps ) , Morbillivirus ( measles ) , Varicelovirus ( chickenpox ) , and others , penetrate their hosts through disruptions of epithelial surfaces and disseminate stepwise to distant organs through a lympho-hematogenic ( LH ) route [2] , [3] . In these cases , we do not know whether PAMPS or T1-IFNs produced at the initial sites of a viral infection can respectively stimulate T1-IFN or ISGs systemically to protect organs before the arrival of the virus or whether the induction and effects of T1-IFNs require local viral replication in the target organ . To counteract the anti-viral effects of T1-IFNs , OPVs including ectromelia virus ( ECTV , the causative agent of mousepox ) , variola virus ( VARV , the agent of smallpox ) , monkeypox virus ( MPXV , the agent of monkeypox ) and vaccinia virus ( VACV , the virus in the smallpox vaccine ) encode a highly conserved T1-IFN binding protein ( T1-IFN bp ) , an early protein that functions as a decoy to divert T1-IFN from the cellular receptor [4]–[9] . Despite that the ECTV T1-IFNbp blocks mouse IFN-α but not IFN-β [5] , it is essential for its virulence [10] . Still , how and where the T1-IFNbp exerts its effects in vivo is not known . It is generally assumed that the major mechanism whereby antibodies protect from viral diseases in general and OPVs in particular is through viral particle neutralization . Alternatively , Ab protection may results from Ab effector functions such as the induction of antibody dependent cellular cytoxicity ( ADCC ) , the promotion of phagocytosis and the activation of the complement cascade to eliminate virions and/or infected cells [11]–[13] . It is well established that Abs that block secreted bacterial virulence factors such as the toxins produce by Clostridia are protective [14] . Some viral immune evasion molecules , including the T1-IFNbp of OPVs , are secreted and theoretically similarly susceptible to the action of Abs [15] . Whether Abs that block the function of these virulence factors can protect or cure viral diseases is not known . If they do , they could provide new opportunities for anti-viral intervention . We have recently shown that ECTV T1-IFNbp induces antibody ( Ab ) responses during infection and that , despite being an non-structural protein , immunization with recombinant T1-IFNbp protects mice from mousepox [10] . However , the mechanism of this protection remains undefined . The pathogenesis of ECTV serves as the classic textbook example of stepwise pathogenesis [3] , [16] . ECTV infects through microabrasions in the footpad , spreads via draining lymph nodes ( D-LN ) and the blood to infect the liver and spleen , and causes death 8–11 days post infection ( dpi ) due to acute liver failure [17] . Here we used ECTV as a model to show that local as opposed to distant infection mediates T1-IFNs production and ISG induction during infection with a virus that disseminate following the common LH route . Moreover , we demonstrate that the T1-IFNbp exerts its effects by attaching to uninfected cells p to block T1-IFN signaling . Finally , we show that Abs that block the biological activity of the T1-IFNbp cure mousepox late in infection demonstrating for the first time that Abs to a secreted immune evasion protein can cure a viral disease .
To determine when T1-IFN and ISG are induced during ECTV stepwise dissemination , we determined T1-IFN ( IFN-β and IFN-α5 ) and ISG ( Mx1 , IRF-7 and sometimes ISG15 ) transcripts in organs of ECTV infected or uninfected BALB/c mice by quantitative PCR ( qPCR ) . Preliminary experiments indicated these T1-IFNs and ISGs are representative of several other T-1IFNs and ISGs . We focused on the popliteal D-LN because it is an obligatory D-LN for ECTV spread , and on the liver , because it is the major target organ of ECTV and liver necrosis is thought to be the cause of death during acute mousepox . At 3 dpi with ECTV , transcripts for T1-IFNs and ISGs increased in the D-LN as compared to uninfected ( 0 dpi ) mice ( Figure 1A ) and virus titers were 6 . 142±0 . 1 Log10 PFU/organ . At this early time point , T1-IFN and ISG transcripts had not been induced in the liver ( Figure 1B ) . Furthermore , T1-IFN was not detected in the serum using a sensitive biological assay ( Figure 1C ) . This indicated that virus replication and T1-IFN production in the D-LN did not result in systemically available T1-IFN or in T1-IFN production or IFN signaling in the liver . The appearance of ISGs ( Figure 1B , right panel ) in the liver followed the appearance of virus and T1-IFN transcripts in the organ ( Figure 1B , left panel ) indicating that local virus replication and T1-IFN production are respectively required for T1-IFN and ISG induction in the liver . Similar to T1-IFN transcripts , ISG transcripts increased in the liver from 3 to 5 dpi . However , while virus loads and T1-IFN transcripts continued to increase from 5 to 7 dpi , ISG transcripts decreased ( Figure 1B ) suggesting a blockade of T1-IFN signaling in the liver of infected mice . In addition , significant T1-IFN activity was observed in the serum at 5 and 7 dpi ( Figure 1C ) . This was IFN-β because the biological assay was inhibited by pre-treating the serum with anti-IFN-β but not with anti-IFN-α Ab ( not shown ) . The absence of IFN-α activity could be due to the action of the T1-IFNbp . We used a sandwich ELISA to test whether at different dpi , T1-IFNbp was released systemically ( Figure 1C ) . At 3 dpi , the amount of T1-IFNbp in sera was variable ( 55±37 ng/ml ) but significantly different from uninfected mice ( P = 0 . 0251 ) indicating that the soluble protein percolated systemically . The amount of T1-IFNbp did not increase significantly between 3 and 5 dpi ( 209±57 ng/ml ) . However , there was a highly significant 20 fold increase ( P = 0 . 0001 ) from 5 to 7 dpi ( 4 , 291±1 , 172 ng/ml ) . To directly determine whether the T1-IFNbp was responsible for the decrease in ISG transcription in the liver at 7 dpi , we used ECTV Δ166-GFP ( a mutant where EVM166 , the gene coding for the T1-IFNbp , was replaced by green fluorescence protein ) which is highly attenuated in BALB/c mice [10] but is lethal to severe combined immunodeficient ( SCID [18] ) mice . As in BALB/c mice infected with WT ECTV , the appearance of T1-IFNs ( Figure 1E , left panel ) and ISG ( Figure 1E , right panel ) transcripts in the liver of SCID mice also followed the appearance of virus . Thus , the relatively low levels of T1-IFNbp in the blood during WT ECTV infection do not appear to be responsible for the lack of systemic induction of ISG in the liver . Most likely , this is because unlike other frequently used models such as intraperitoneal infection with lymphocytic choriomeningitis virus ( LCMV , Figure 1C , clear circles ) there is no systemic T1-IFN activity during early ECTV infection . Still , different to WT ECTV infection of BALB/c mice , the transcription of ISG in the liver of SCID mice infected with ECTV Δ166 increased significantly between 5 and 7 dpi ( Figure 1E , right panel ) . Consequently , even though systemic T1-IFN activity was present at 7 dpi , the T1-IFNbp was able to dampen T1-IFN signaling in the liver of BALB/c mice infected with WT ECTV . In vitro , the early T1-IFNbp is secreted from infected cells , but can also bind back to the surface of infected and uninfected cells [8] , [19] by attaching to glycosaminoglycans at the cell membrane [20] . Consistent with these reports , when we infected L cells with 1 MOI ECTV-GFP ( a recombinant ECTV expressing non-structural cytosolic GFP under the early/late 7 . 5 VACV promoter [21] and as virulent as WT ECTV , [10] ) , both , the GFP+ and GFP- cell populations were stained by the anti-T1-IFNbp sera . Presumably , the GFP- cells were not actively synthesizing viral proteins and most likely acquired T1-IFNbp from the GFP+ cells . As a control , L cells infected with ECTV Δ166-GFP did not stain with anti- T1-IFNbp ( Figure 2 ) . We tested whether a similar phenomenon occurred in vivo . At 3 dpi , very few cells stained with rabbit antisera toT1-IFNbp or to EVM135 , the ECTV ortholog of the early/late VACV structural protein A33R [9] , [22] which exclusively stains infected cells . On the other hand , at 5 dpi , both antisera stained numerous foci of cells in serial sections of the livers of BALB/c mice . While coincident in space , the foci stained with anti-T1-IFNbp were significantly larger than those stained with anti-EVM135 suggesting that the T1-IFNbp spread further than the virus itself ( Figure 3A ) . Consistent with the large quantities of T1-IFNbp in the serum , at 7 dpi , the liver appeared saturated when stained with anti-T1-IFNbp even though some areas did not stain with anti-EVM135 ( Figure 3B ) . We also infected mice with ECTV expressing cytosolic firefly luciferase ( ECTV-Luc ) controlled by the 7 . 5 promoter as a surrogate of viral protein . This allowed us to perform two-color immunoflourescence using rabbit anti-T1-IFNbp and goat anti-Luc to reveal T1-IFNbp bound to uninfected cells surrounding liver infected foci at 5 and 7 dpi ( Figure 3C ) . A similar phenomenon was also observed in the spleen , which is also a target of ECTV ( Figure S1 ) . Thus , as in tissue culture , secreted T1-IFNbp binds to the surface of uninfected cells in vivo . It was of interest to test whether T1-IFNbp Ab could also bind to the surface of cells in and surrounding infected foci in the liver following in vivo administration . For this purpose , BALB/c mice infected with ECTV-Luc were given 200 µl anti-T1-IFNbp or control naive sera at 5 dpi and their livers stained with anti-rabbit and anti-luc 16 h later . We found that the cells within and surrounding infected foci in the liver were decorated with rabbit IgG in mice treated with T1-IFNbp antisera but not with control sera ( Figure 4A ) . Given that T1-IFNbp Ab bound to infected foci in vivo , we tested whether T1-IFNbp antisera could prevent or cure mousepox . We inoculated BALB/c mice with rabbit T1-IFNbp antisera on different days post infection ( dpi ) . The antisera significantly protected from lethality when given as late as at 5 dpi ( Figure 4B ) . Treatment with T1-IFNbp antisera at 5 dpi significantly reduced virus titers in the liver ( Figure 4C ) and liver necrosis seen as dark pink areas devoid of nuclei ( Figure 4D ) at 2 days post treatment ( dpt ) . The protection observed with the T1-IFNbp antisera could be due to an ability to restore T1-IFN signaling ( i . e . by inhibiting the biological activity of the T1-IFNbp ) and/or by traditional Ab effector mechanisms such as antibody dependent cellular cytotoxicity ( ADCC ) , phagocytosis , or complement activation [11]–[13] . We identified two mAbs , 10F3 and 10G7 that bound to recombinant ECTV T1-IFNbp with similar efficiency in ELISA ( Figure 5A ) and at the surface of cells incubated with recombinant ECTV T1-IFNbp ( Figure 5B ) or infected with ECTV-GFP ( Figure 5C ) . However , 10G7 fully blocked the ability of ECTV T1-IFNbp to inhibit the antiviral function of mouse IFN-α as determined in vesicular stomatitis virus ( VSV ) inhibition assays while 10F3 had a very moderate inhibitory effect ( Figure 5D ) . Both mAbs were IgG1 , an isotype known to have poor effector function in the mouse [23] , [24] . To determine whether 10G7 and 10F3 also differed in their ability to inhibit the biological function of T1-IFNbp in vivo , BALB/c mice were infected with ECTV and at 5 dpi treated with inhibitory 10G7 or poorly-inhibitory 10F3 . At one dpt , the virus titers ( not shown ) and T1-IFN transcripts in the livers of 10G7- and 10F3-treated mice were similar . However , ISG transcription in mice treated with inhibitory 10G7 was significantly increased as compared to mice treated with 10F3 ( Figure 6A ) . In other experiments , mice treated with isotype control also failed to upregulate T1-IFN and ISG transcripts ( not shown ) The upregulation of ISGs after 10G7 treatment was not due to direct stimulation of IFNAR by 10G7 because uninfected mice treated with 10G7 did not upregulate Mx1 ( Figure S2 ) . Moreover , phosphorylated Stat1 in the nuclei of infected hepatocytes was readily apparent in the livers of mice treated with 10G7 but not with 10F3 ( Figure 6B ) . At 2 dpt , mice treated with inhibitory 10G7 had significantly lower virus titers in their livers as compared with mice treated with inhibitory 10F3 or isotype control in plaque assays ( Figure 6C ) suggesting that the increase in ISG transcription resulted in improved antiviral state . Moreover , immunohistochemistry with anti-EVM135 sera at 2 and 3 dpt ( Figure 6D upper and middle panels and Figure 6E ) ) showed that the foci in mice treated with inhibitory 10G7 were smaller than those of mice treated with 10F3 or IC suggesting reduced virus spread . Also , at 3 dpt the livers of mice treated with inhibitory 10F3 or IC were necrotic but not those from mice treated with inhibitory 10G7 ( Figure 6D , lower panels ) . Finally , the survival of mice treated with inhibitory 10G7 was significantly higher than of those treated with 10F3 or IC ( Figure 6F ) . Thus , while 10F3 and 10G7 T1-IFNbp mAbs bound equally well to cell surfaces and similar numbers of inflammatory cells were recruited to the livers of treated mice , only the inhibitory mAb 10G7 rescued T1-IFN signaling , decreased liver damage and virus loads , and prevented lethal mousepox . Decreased virus loads after 10G7 treatment was also observed when mice were infected with ECTV-Luc and the virus loads visualized by whole animal imaging ( Figure 7 ) . Additionally , at 2 dpt significantly more leukocytes and more CD8+ T cells were recovered from the livers of mice treated with inhibitory 10G7 and 10F3 as compared with uninfected mice . On the other hand , no significant differences were observed between groups of mice treated with 10G7 , 10F3 or IC . Also , no significant differences were observed in other immune populations ( Figure S3 ) . Thus , mAb treatment did not affect the infiltration of leukocytes in the liver that normally occurs during ECTV infection . Given that 10G7 can inhibit the biological activity of T1-IFNbp from ECTV and protect from disease , it was of interest to determine whether 10G7 can also block the biological activity of T1-IFNbp from OPVs important to human health . We found that 10G7 recognized cells pre-incubated with recombinant T1-IFNbp from the OPV variola virus ( VARV , the agent of human smallpox ) , or supernatants of cells infected with the OPVs vaccinia virus ( VACV , the virus in the smallpox vaccine ) or monkeypox virus ( MPXV , endemic in human populations of central Africa ) ( Figure 8A ) . Furthermore , while the antiviral activity of 10 U/ml Human IFN-α ( hIFN-α ) was inhibited by recombinant VARV T1-IFNbp or supernatants of MPXV infected cells ( [4] , [6] , [19] , [25] Figure 8B ) ; pre-incubation of VARV T1-IFNbp or MPXV supernatants with mAb 10G7 blocked their ability to inhibit the antiviral function of hIFN-α ( Figure 8C ) demonstrating the ability of 10G7 to block the biological function of T1-IFNbp from different OPVs .
The results presented here provide novel information regardingT1-IFN induction and protection in vivo . While the induction of T1-IFNs in vivo following systemic ( intravenous ( i . v . ) or intraperitoneal ( i . p . ) administration of viruses has been studied to a great extent [26]–[28] and their ability to induce an antiviral response is well known [1] , we still lack an understanding of how T1-IFNs and ISGs are temporally induced and protect from disease during the course of the many viral infections that follow a stepwise mode of LH dissemination [16] . The induction of T1-IFN genes depends on cells sensing viral infection . Cells recognize Pathogen Associated Molecular Patterns ( PAMPs ) of viruses ( in most cases nucleic acids ) by means of Pathogen Recognition Receptors ( PRR ) expressed at the plasma membrane ( e . g . Toll Like Receptor ( TLR ) 2 , TLR4 ) , in endosomes ( TLR3 , TLR7 , TLR9 ) or in the cytosol ( RIG-I , MDA5 , DAI ) . [29]–[31] . Signaling through PRRs culminate in the activation of specific members of the IRF family of transcription factors , most notably IRF3 , IRF7 , NF-κB and c-jun which stimulate the T1-IFN promoters . During stepwise infection , T1-IFNs could act on vital target organ indirectly . For example , they could induce ISGs and help orchestrate the innate and adaptive immune response in the D-LN thereby curbing virus spread to the target organ . As we have previously shown , this is a major mechanism whereby NK [32]–[34] and memory CD8+ T cells [35] protect mice from mousepox . Alternatively , T1-IFNs could directly induce ISGs in the target organ and/or contribute to the recruitment of immune cells . In this case , the protection of the target organ could result from the T1-IFN produced at the primary site of infection that is distributed systemically , or from the T1-IFN produced locally in the target organ from PAMPS either distributed systemically or locally produced . Here we have used the classical ECTV model of LH spread to show that during ECTV infection , T1-IFN signaling in the liver ( the target organ ) strongly correlates with resistance to disease . We also show that ISG induction in the liver correlates with T1-IFN transcription in the liver but not in the D-LN suggesting that T1-IFN signaling in the liver is reliant on local T1-IFN production . Moreover , we demonstrate that the induction of T1-IFN in the liver depends exclusively on local viral replication . This suggests that PAMPS produced in the footpad or in the D-LN do not distribute systemically to the liver before virus arrival . It should be noted , however , that poxviruses excel in the number of immune evasion proteins that affect innate immunity [36] . Hence , it is possible that during other viral infections where T1-IFN production and signaling is not targeted by the virus , the systemic distribution of T1-IFN may have a more important role in distant ISG induction . Our work also impinges on our understanding of viral immune evasion . During the past few years there has been much progress towards the characterization of virally encoded immune evasion genes . While the cellular and molecular mechanisms whereby many of these evasion molecules operate are well known [36]–[39] , we still have an incomplete understanding on how they subvert the immune response in vivo . We have previously shown that the OPV T1-IFNbp is secreted from infected cells and binds back to cell surfaces [19] by attaching to glycosaminoglycans at the cell membrane [20] . Whether this also occurs in vivo and is significant for viral virulence remained unknown . Our experiments reveal that the T1-IFNbp produces its evasive effect at least in part by attaching to uninfected liver cells surrounding infected foci , thereby precluding their ability to signal through the T1-IFN receptor . Of interest , the ECTV T1-IFNbp does not block IFN-β although ECTV infection induced IFN-β transcription . A remaining question is why IFN-β was insufficient to induce high levels of ISGs; our data may indicate that in vivo , IFN-α and IFN-β have different functions . Our results also have implications to our understanding of Ab-mediated protection from viral disease . The most commonly accepted mechanism of Ab protection is viral particle neutralization [11]–[13] . Indeed , it has been suggested that this is the mechanism whereby the smallpox vaccine protects [40] . However , while clinical data showed that protection from smallpox correlated with Ab neutralization , the same investigators could not find a causal association between neutralizing Ab titers and protection against smallpox [41] , [42] . These findings suggests that mechanisms other than viral particle neutralization may be involved in protection by the smallpox vaccine . In support of this , Benhnia et al . recently demonstrated that mice can be protected from VACV by prior administration of Abs to the structural protein B5R and that this protection relied in complement activation [13] , [43] , a well known Ab effector mechanism . Here we show that Abs can protect from and cure advanced systemic OPV disease by a previously unsuspected mechanism: inhibiting the function of an immune evasion protein . Of interest , while it is known that polyclonal antibodies in the form of convalescent sera and vaccinia immunoglobulin ( VIG ) are protective pre- and soon after exposure to VARV , the possible mechanisms of this protection remain unknown [12] , [13] . We show that Abs to the T1-IFNbp cure mousepox even when administered as late as at 5 dpi suggesting that Abs to the T1-IFNbp may play a role in protection not only by the smallpox vaccine but also by VIG because , at least in mice , anti T1-IFNbp are present in sera following VACV vaccination [10] . Thus , our experiments uncovered a novel mechanism of Ab mediated protection . We have previously shown the importance of the T-1 IFNbp in ECTV pathogenesis [10] . Whether Abs to other secreted immunoregulatory viral protein could have a similar effect will likely depend on whether they play an essential role in pathogenesis or not and remains to be studied . It is interesting to note that while resistance to primary infection with ECTV requires T1-IFN function , resistance to secondary infection does not as IFNAR1 deficient mice immunized with attenuated ECTV or VACV resisted a challenge with WT ECTV [44] , [45] . This suggests that the main role of T1-IFNs in protection is to control the virus until an adaptive response is generated and thereafter become irrelevant . Hence , it is likely that Abs to the T1-IFNbp also have a similar effect . Drugs being tested for the treatment of OPV infections are ST-246 [46] , which targets VACV F13L protein and its orthologs in other OPVs to inhibit the egress of extracellular virions from cells , and CMX001 , an oral ether-lipid analogue of the acyclic nucleoside phosphonate Cidofovir [47] , that target the viral DNA polymerase . These two drug types have been very effective for the treatment of various OPVs in several animal models [48]–[54] . CMX001 has been shown to cure intranasal ECTV infection when treatment was started as late as at 5 dpi [51] . Still , there is the caveat that OPVs could naturally develop resistance , or that resistant viruses could be artificially created . Indeed , VACV resistant to Cidofovir and its derivatives has been demonstrated [55]–[59] . Similarly , cowpox virus resistant to ST-246 has been isolated [46] . Thus , more than one or two anti-poxvirus drugs directed towards different targets are needed . 10G7 mAb or similar T1-IFNbp mAbs could be exploited to treat humans against OPV infections because OPVs that affect humans encode a T1-IFNbp and at least three of them ( VACV , VARV and MPXV ) are inhibited by 10G7 . In addition , a strategy of using mAbs to inhibit secreted immune evasion proteins important for viral pathogenesis could be explored to prevent and treat infections with any other virus that encode proteins of this kind .
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 . All protocols were approved by Fox Chase Cancer Center's Institutional Animal Care and Use Committee . BALB/c female mice were purchased from Taconic Farms . SCID mice in a BALB/c background were bred at FCCC Laboratory Animal Facility . All mice used in experiments were 5–12 week old . Media and cells were as previously described [10] , [60] , [61] . For antibody production , hybridomas were grown in a Celline devise ( BD ) using protein free MAb Medium ( BD ) as recommended by the manufacturer . The mAbs were purified from the medium using ammonium sulfate precipitation and purity confirmed by SDS-PAGE gel . Stocks of ECTV Moscow strain ( ATCC VR-1374 ) were propagated in tissue culture as previously described [10] . Virus titers were performed in BS-C-1 cells . Stocks of VSV Indiana VSV-eGFP were a gift of Dr . S . Balachandran . They were expanded in BHK cells and virus titers were determined in Vero cells by plaque assay as previously described [10] . For the generation of recombinant ECTV expressing firefly luciferase ( Luc , GenBank accession number AAL30790 ) , we used homologous recombination as previously described using ECTV-GFP [10] as host virus to replace GFP for Luc . Non-fluorescent plaques were purified 10 times , expanded , and the insertion sequenced with primers flanking the site of homologous recombination . This virus was as pathogenic as WT virus in LD50 experiments . Mice were infected in the footpad with 100 PFU ECTV WT or Luc as indicated . For the determination of survival , the mice were monitored daily . To avoid unnecessary suffering , mice were euthanized and counted as dead if imminent death was certain . For virus titers mice were infected with 100 PFU ECTV . Mice were euthanized when indicated and whole LNs or 100 mg of liver were homogenized in PBS using a Tissue Lyser homogenizer ( Qiagen ) . Virus titers were determined on BS-C-1 cells in 6-well-plates as before [10] , [60] , [61] . MPXV RoC ( 2003 358 ) and MPXV USA ( 2003 044 ) ( Center for Disease Control , USA ) were incubated on a BSC-40 monolayer ( 7 . 7×106 cells ) at a MOI of 1 for one hour at 36 oC , 6% CO2 in 1 ml of Opti-MEM . After one hour the infectious inoculum was removed , replaced with 2 ml Opti-MEM and incubated for 18 hours . The supernatants were collected and spun at 1000×g for 10 min . in a tabletop microcentrifuge to remove cell debris . After transfer to a new microcentrifuge tube , the supernatants were stored at −70 oC . The samples were subjected to 4 . 4×106 rads of gamma rays for 4 hours on dry ice and stored at −70°C until use . Production of recombinant T1-IFNbp from ECTV and VARV was exactly as described previously [6] , [10] . Rabbits were immunized three times at 1 month interval with recombinant T1-IFNbp [10] or EVM135 [62] in incomplete Freund's adjuvant , sera were obtained 1 month after last injection . Antisera were evaluated for antibodies by ELISA . To generate mouse hybridomas producing T1-IFNbp specific mAbs , BALb/c mice were immunized three times with 50 µg recombinant T1-IFNbp s . c . in incomplete Freund adjuvant . After one month of rest , the mice were boosted with 10 µg T1-IFNbp in PBS intravenous and their spleens fused the next day using standard hybridoma procedures at the FCCC Hybridoma Facility . The initial screening of mAb against T1-IFNbp was performed by ELISA . Positive mAbs were further analyzed for their ability to block T1-IFNbp using a VSV inhibition assay . To compare the binding of Binding of mAbs to T1-IFNbp the T1-IFNbp was determined by ELISA assay was performed with immobilized T1-IFNbp as previously [10] but using dilutions of mAbs rather than serum . For the detection of T1-IFNbp in serum , we used a sandwich ELISA . For this purpose , high-binding 96-well plates ( Corning ) were coated with mouse anti-T1-IFNbp ( 10G7 , 50ng/well ) at 4°C overnight . After washing and blocking , diluted sera ( 1∶10 dilution in PBS ) were added to each well , and incubated at 37°C for 1 hr . After washing , secondary rabbit anti-T1-IFNbp serum ( 1∶1000 in PBS ) was added and incubated at RT for 2 hr . The plates were washed four times and incubated with HRP conjugated anti-Rabbit IgG ( KPL , 100 ng/well ) at RT for 1 hr . Following washing , signal was developed with 100 ul of TMB ( Sigma , USA ) and the reaction was stopped by adding 0 . 5N sulfuric acid . The OD was determined at 450 nm using a multiwell plate reader . For quantification , a serial dilution ( 10 ug to 1 pg/ml ) of recombinant T1-IFNbp was included in the same plate . Concentrations of biologically active T1-IFN in serum were measured using an ISRE- ( interferon stimulated responsive element ) luc reporter assay as described [63] . Briefly , 1∶10 diluted sera were overlaid on L929 ISRE-Luc reporter cells ( a gift from Dr . Russell Vance , University of California , Berkeley ) in a 96-well plate and incubated overnight at 37°C . L929 ISRE-Luc reporter cells were lysed and the luciferase activity was measured by adding firefly luciferin substrate ( Agilent Technologies ) and measuring luminescence in a 96-well plate reader . Recombinant IFN-β ( PBL Interferon Source ) was included in the same plate for quantification . The VSV inhibition assays were modified from those described before [10] . Briefly , 10 µl of mIFN-α ( 0 . 1 I/ul , PBL ) was incubated or not with recombinant T1-IFNbp ( 10 ng/ml ) or with T1-IFNbp and the indicated mAbs ( 100 µl of supernatant ) for 1 h at 37°C . IFN-α or the indicated mixtures were then added to L929 cells in a 12-well-plate and incubated at 37°C for 24 h to induce ( or not ) an antiviral state . The cells were then infected with VSV-GFP ( at MOI of 0 . 01 for 16 h ) or VSV ( at MOI of 0 . 01 for 48 h ) . Protection or lack of protection of the cells by the mIFN-α was assessed by determining the expression of GFP under fluorescent microscope or by staining with crystal violet . The blockade of the VARV and MPXV T1-IFNbps by the mAbs were determined similarly but using HeLa cells in 24-well plates instead of L929 cells in 12 well plates , recombinant human IFN-α ( PBL ) instead of mouse IFN-α , and recombinant VARV T1-IFNbp or irradiated supernatant of BS-C-1 cells that had been infected with the indicated strains of MPXV instead of ECTV T1-IFNbp . In this case , protection of the cells was determined by staining with crystal violet . To detect T1-IFNbp expression at the cell surface , 106 L929 cells were infected with ECTV-GFP or ECTV-Δ166-GFP at MOI of 1 for 24 h , the next day , the cells were trypsinized and incubated with 100 µl of 1/1000 T1-IFNbp antisera or naïve control serum , or incubated with 10 ng of mAbs ( 10F3 and 10G7 ) in 100 µl PBS for 1 h at 37°C . The cells were then washed , stained with PE-conjugated goat anti-rabbit IgG Ab or PE-conjugated goat anti-mouse IgG Ab for 30 min , respectively and analyzed using an LSRII flow cytometer ( BD ) . For detection of mAb binding to recombinant T1-IFNbp attached to the surface of uninfected cells , L929 cells were incubated with 10 ng of recombinant T1-IFNbp at 37°C for 1 h , washed , incubated with 10 ng of mAb ( 10 F3 or 10 G7 ) for 1h , washed again and stained with PE goat anti-mouse IgG Ab . For the detection mAb binding to recombinant VARV T1-IFNbp [6] and supernatant of MPVX infected cells , HeLa cells were incubated with 10 ng of recombinant VARV T1-IFNbp or 100 ul of irradiated supernatant from BS-C-1 cells that been infected with the indicated strains of MPVX for 1 h at 37°C , washed , incubated with 10 ng of mAb ( 10 F3 or 10 G7 ) for 1 h at RT , washed again and stained with PE conjugated goat anti-mouse IgG Ab for 30 min washed and analyzed by flow cytometry . For the analysis of splenocytes and liver infiltrating leukocytes , BALB/c mice were infected with 100 PFU of ECTV . At 5 dpi , 200 ug of indicated mAb were injected i . p . 2 dpt , mice spleens and livers were harvested . Spleens were made into single cell suspensions and infiltrating lymphocytes were isolated from the livers using a Percoll gradient as previously described [64] . Cells were surface stained with Cy7PE-anti-CD8a , APC-anti-CD49b , PE-anti-CD4 and Cascade Blue anti-CD3 ( Biolegend ) and intracellularly with PE-anti-IFN-γ ( Biolegend ) and cy5 . 5PE-anti human Granzyme B ( Invitrogen ) as described previously [10] , [60] , [61] . Flow cytometry was performed using an LSRII flow cytometer ( BD ) and Flowjo software for analysis . These were performed by standard procedures . Briefly , mouse livers were fixed with formalin and embedded in paraffin . 5 µM sections were stained with rabbit anti-EVM135 or T1-IFNbp antisera diluted 1∶500 , followed by biotin goat anti-rabbit and extravidin-peroxidase each for 30 min . The slides were revealed using chromogen-DAB ( Sigma ) as substrate and counter stain with Gill's Hematoxylin . For double immunohistochemistry staining , BALB/c mice were infected with ECTV-luc for 5 days; liver sections were prepared and stained with the above procedure in two steps . In the first step , sections were stained with goat anti-Luciferase Ab and anti-goat IgG-HRP secondary and chromogen-DAB ( Sigma ) as the substrate . In the second step , sections were stained with rabbit antiserum to T1-IFNbp and anti-Rabbit IgG-HRP , the substrate was VIP ( Vector Lab SK-4600 ) . For immunofluorescent double staining of luciferase and T1-IFNbp , mice were infected with ECTV-Luc rather than WT ECTV . Livers were immersed in PBS and snapped frozen in liquid nitrogen . 10 µM cryosections were fixed in 95% acetone at −20°C for 10 min , air dried at room temperature , and sequentially stained with goat anti-luciferase ( Invitrogen ) ( 1∶400 ) , rabbit anti-T1-IFNbp antiserum ( 1∶400 ) for 60 min at RT , anti-goat Alexa Fluor 488 and anti-Rabbit Alexa fluor 555 ( Invitrogen ) at 1∶500 dilution for 60 min . after several washes , slides were mounted under coverslips with one drop of Fluoromount G . For immunofluorescent double staining of luciferase and phosphor-Stat1 , mice were infected with 1000 PFU ECTV-luc and treated at 5 and 6 dpi with 200 µg of the indicated mAbs . At 7 dpi , the livers were fixed with 4% paraformaldehyde for 30 minutes , washed and permeabilized with 1% Triton X-100 . Slides were stained with goat anti-luciferase and rabbit anti-phospho Stat1 overnight . After several washes cells were stained with donkey anti-goat-DyLight 488 and donkey anti-rabbit-Alexa fluor 647 ( both from Jackson ImmunoResearch ) for 90 minutes , and then washed before mounting in Fluromount containing DAPI . For detection of distribution of rabbit Ab in vivo , BALB/c mice were infected with ECTV-Luc in the footpad . At 5 dpi , 200 µl of antisera were injected i . p . , 16 h later , mice were sacrificed and livers were snap frozen and cut into 10 um sections . Following fixation , sections were stained with Alexa-555 conjugated anti-rabbit IgG ( Invitrogen ) and goat anti-luciferase for 30 min and Alexa-488 conjugated anti-goat IgG Ab for another 30 min . In vivo bioluminescent imaging was performed using a Carestream In-Vivo Multispectral FX PRO Imaging System ( Carestream healthcare ) . Briefly , BALB/c mice were infected with 300 PFU of ECTV-luc in the footpad , at 5 dpi , mice were treated with indicated Ab i . p . , 2 dpt , the mice were anesthetized using ketamine ( 70 mg/kg of body weight ) and xylazine ( 7 mg/kg of body weight ) and 150 mg/kg of D-luciferin substrate was administered i . p . exactly 10 min before acquisition . Luminescence was captured with an exposure time of 10 s . Five mice were imaged at each time . The mean luminescent intensity was determined using Carestream's Molecular Imaging software . RNA was isolated from organs using Trizol reagent ( Invitrogen ) according to manufacturer's instructions . Total RNA was treated with DNase I ( Qiagen ) and further purified using the RNeasy Mini Kit ( Qiagen ) . 2 µg of total RNA samples were reverse transcribed using the High Capacity cDNA Reverse Transcription Kit ( Applied Biosystems ) . 1 ng cDNA was amplified by real time PCR using TaqMan Probes for ifna5 ( ID:Mm00833976_s1 ) , ifna2 ( ID:Mm00833961_s1 ) , ifnb1 ( ID:Mm00439546_s1 ) , Mx1 ( ID:Mm00487796_m1 ) and IRF7 ( ID:Mm00516793_g1 ) , and GAPDH ( ID:Mm99999915_g1 ) as an internal control for normalization . Each sample was run in 20 ul reaction using TaqMan Universal PCR Master Mix . Reactions were performed in an ABI real time PCR 7500 ( Applied Biosystems , Foster City , CA ) . Ratios of mRNA levels to control values were calculated using the ΔCt method ( 2-ΔΔCt ) at a threshold of 0 . 02 [65] . All data were normalized to control GAPDH . PCR conditions used: hold for 10 min at 95°C , followed by 40 cycles of 15 s at 95°C and 60 s at 60°C . Statistical analyses were performed using Prism software . | Type 1 interferons are molecules important in the defense against viruses . Orthopoxviruses encode a Type 1 interferon binding protein that acts as a decoy for the Type 1 interferon receptor . Here we show that during infection with the Orthopoxvirus ectromelia virus , the agent of mousepox , Type 1 interferons protect the liver locally rather than systemically . We also show that the Type 1 interferon binding protein of ectromelia virus attaches to uninfected cells surrounding infected foci in the liver to impair their ability to receive Type 1 interferon signaling and facilitate virus spread and disease progression . We also show that this process can be reversed and mousepox cured late in infection by treating mice with antibodies that block the biological function of the Type 1 interferon binding protein . Because the Type 1 interferon binding proteins of different Orthopoxviruses are very well conserved , the antibodies also block the biological function of the Type 1 interferon binding proteins from variola virus ( the virus of smallpox ) and monkeypoxvirus . Thus , our findings provide insights on how Type 1 interferons function and are evaded during a viral infection in vivo , and unveil a novel mechanism for antibody-mediated antiviral therapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunity",
"to",
"infections",
"immunology",
"microbiology",
"host-pathogen",
"interaction",
"mechanisms",
"of",
"resistance",
"and",
"susceptibility",
"immunotherapy",
"animal",
"models",
"of",
"infection",
"virulence",
"factors",
"and",
"mechanisms",
"viral",
"immune",... | 2012 | Antibody Inhibition of a Viral Type 1 Interferon Decoy Receptor Cures a Viral Disease by Restoring Interferon Signaling in the Liver |
Two distinct Polycomb complexes , PRC1 and PRC2 , collaborate to maintain epigenetic repression of key developmental loci in embryonic stem cells ( ESCs ) . PRC1 and PRC2 have histone modifying activities , catalyzing mono-ubiquitination of histone H2A ( H2AK119u1 ) and trimethylation of H3 lysine 27 ( H3K27me3 ) , respectively . Compared to H3K27me3 , localization and the role of H2AK119u1 are not fully understood in ESCs . Here we present genome-wide H2AK119u1 maps in ESCs and identify a group of genes at which H2AK119u1 is deposited in a Ring1-dependent manner . These genes are a distinctive subset of genes with H3K27me3 enrichment and are the central targets of Polycomb silencing that are required to maintain ESC identity . We further show that the H2A ubiquitination activity of PRC1 is dispensable for its target binding and its activity to compact chromatin at Hox loci , but is indispensable for efficient repression of target genes and thereby ESC maintenance . These data demonstrate that multiple effector mechanisms including H2A ubiquitination and chromatin compaction combine to mediate PRC1-dependent repression of genes that are crucial for the maintenance of ESC identity . Utilization of these diverse effector mechanisms might provide a means to maintain a repressive state that is robust yet highly responsive to developmental cues during ES cell self-renewal and differentiation .
Embryonic stem cells ( ESCs ) can undergo unlimited self-renewal while maintaining their pluripotent and undifferentiated states , features that are consistent with their origin within the inner cell mass of the blastocyst . Increasing evidence suggests that in addition to the core gene regulatory circuitry composed of Oct3/4 , Sox2 , Nanog and other transcription factors , Polycomb group proteins critically contribute to maintain the undifferentiated state of ESCs by silencing genes that are involved in development and/or transcription [1] , [2] , [3] , [4] , [5] , [6] . Polycomb-mediated repression of these genes is also essential to preserve the ability of ES cells to differentiate in response to extracellular cues [7] , [8] , [9] . Polycomb group proteins are chromatin-modifiers that mediate transcriptional repression . They form at least two types of multimeric complexes , the Polycomb repressive complexes-1 ( PRC1 ) and -2 ( PRC2 ) , the core components of which are conserved from Drosophila to human [10] , [11] , [12] , [13] , [14] . PRC2 contains Ezh2 or -1 , which catalyze trimethylation of histone H3 lysine 27 ( H3K27me3 ) , a posttranslational modification that is thought to be recognized by the chromo-domain ( CHD ) protein components of PRC1 [12] , [13] , [14] , [15] , [16] . Within PRC1 , Ring1B and –A act as major E3 ubiquitin ligases for histone H2A mono-ubiquitination at lysine 119 ( H2AK119u1 ) [17] , [18] . Conditional depletion of Ring1B and –A in ESCs leads to global loss of H2AK119u1 and concurrent derepression of ‘bivalent’ genes enriched for both H3K27me3 and H3K4me3 [5] , [19] . H2AK119u1 deposition has been shown to localize to the inactive X chromosome ( Xi ) , the XY body , and several silenced ‘bivalent’ loci in mouse ESCs [19] , [20] , [21] . Recent genome-wide H2AK119u1 analysis in MEFs ( mouse embryonic fibroblast ) has revealed Bmi1-dependent deposition of H2AK119u1 at the promoter regions of many repressed genes [22] . These findings suggest that H2AK119u1 could be a part of the regulatory process that is required for PRC1-mediated repression . However , the role of H2AK119u1 in PRC1-mediated repression is still controversial . A recent study has reported that Ring1B can compact chromatin structure of the Hox loci and repress Hox expression independent of its E3 activity [23] . This idea has been supported by a previous study which showed that PRC1 components can compress nucleosomal templates assembled from tail-less histones into a form that is refractory to chromatin remodeling in vitro [24] . This hypothesis , however , needs rigorous validation because this study was performed by using Ring1B single knockout ( KO ) cells , in which Ring1A-catalyzed H2AK119u1 still remained in a lower level [5] , [17] , [20] , [25] . In this experimental setup , Ring1A and associated H2AK119u1 may potentially complement Ring1B-mediated chromatin compaction of Hox genes to mediate their repression . Consistently , ESCs are capable to retain ESC-like morphology and LIF-dependent proliferation upon depletion of Ring1B but not doubly depletion of Ring1B and –A [5] , [9] , [20] , [26] . Therefore , to properly estimate in which extent H2AK119u1 contributes to PRC1-dependent repression in ESCs , and experimental platform that excludes Ring1A is necessary . In this study , we first determined the localization of H2AK119u1 in ESCs by ChIP-on-chip analysis and found the H2AK119u1-bound genes as core targets of PRC1-dependent repression . We further demonstrated that catalytic activity of PRC1 towards H2A is essential for silencing of target loci and maintenance of ESCs . We also found PRC1-mediated H2AK119u1 is complemented by independent functions of PRC1 that contribute to gene silencing and chromatin compaction , most notably at Hox loci . We propose that PRC1 combines diverse effecter mechanisms to mediate robust repression of target genes and stable maintenance of undifferentiated status of ESCs .
Global H2AK119u1 distribution has been reported only for MEFs and the human teratocarcinoma NT2 cell line [22] , [27] , but not for mouse ESCs . We , therefore , used ChIP-on-chip analysis to clarify H2AK119u1 deposition around transcription start sites ( TSS ) in mouse ESCs by using an Agilent mouse promoter array and an E6C5 monoclonal antibody ( mAb ) or a rabbit polyclonal antibody [28] . Ring1A/B-dKO ESCs , in which H2AK119u1 is apparently undetectable , were used as a negative control [5] , [19] . Ring1A/B-dKO ESCs were induced by treating Ring1A−/−;Ring1Bfl/fl;Rosa26::CreERT2 ESCs with 4-hydroxy-tamoxifen ( OHT ) , which rapidly activates CreERT2 and catalyzes loxP recombination at Ring1B locus [5] . Distribution of Ring1B , H3K27me3 and H2A were re-examined to obtain a reference data set . E6C5-ChIP signals at the promoter regions of known target loci , Hoxa9 , Pax9 and Tbx3 show that H2AK119u1 deposition is readily detectable in Ring1A−/− ESCs but not in Ring1A/B-dKO ( dKO ) ESCs ( Figure S1 ) . These results were validated by ChIP-qPCR as shown in Figure S2A . We calculated the averages of E6C5-ChIP signals in Ring1B-bound and –unbound genes and found higher H2AK119u1 deposition in Ring1B-bound genes in Ring1A−/− than in Ring1A/B-dKO ( dKO ) whereas little difference was noted for unbound genes ( Figure 1A; Figure S3A and S3B ) . These data therefore appear to reflect H2AK119u1 deposition that depends on Ring1B . For detailed investigation of the genes that exhibit H2AK119u1 enrichment , we examined the gene-wise distribution of E6C5-ChIP signals after subtraction of the background enrichment value and identified 538 target genes ( Figure 1B; Figure S3C; the list of genes is shown in Table S1 ) . The results with the E6C5 mAb were re-confirmed by using a rabbit antiserum that recognizes H2AK119u1 [28] . With this reagent , 524 genes were found to be bound by H2AK119u1 , and these genes significantly overlapped with those identified by the E6C5 mAb ( Figure S3D; Table S1 ) . Taken together , using the above methods we determined a set of genes in ESCs that have H2AK119u1 deposition around their TSS . We went on to examine the correlation of genes enriched for H2AK119u1 ( H2AK119u1+ ) with those having Ring1B ( Ring1B+ ) and H3K27me3 ( H3K27me3+ ) depositions . We found that genes bound by Ring1B and H3K27me3 identified in this study were significantly overlapped with those reported in previous studies ( Figure S3E , F ) . We identified 1721 and 3686 genes bound by Ring1B and H3K27me3 , respectively , and found H2AK119u1+ genes as a subset of the Ring1B+ genes ( Figure 1B; Figure S3C; Table S1 ) . Since most Ring1B+ genes define a subset of H3K27me3+ genes , H3K27me3+ genes could be subdivided into three distinct layers , H2AK119u1+Ring1B+H3K27me3+ ( Triple positive; TP ) , H2AK119u1-Ring1B+H3K27me3+ ( Double positive; DP ) and H2AK119u1-Ring1B-H3K27me3+ ( Single positive; SP ) . We finally confirmed the quantitative difference of H2AK119u1 level at TP genes against DP or SP genes by ChIP-qPCR analysis at selected genes ( Figure S2A ) . Although we cannot exclude a possibility that we failed to detect a low level of H2AK119u1 at some DP genes , our data demonstrate that H3K27me3+ gene promoters are not uniformly occupied by Ring1B and H2AK119u1 . We investigated functional properties of H2AK119u1+ genes among Polycomb targets . Scattered plot analysis for gene-wise deposition of H3K27me3 and Ring1B revealed that H2AK119u1 targets were significantly enriched among genes that have high levels of both Ring1B and H3K27me3 occupancy ( Figure S4 ) . This suggests that TP genes represent the central targets for Polycomb repression . We compared the impact of PRC1 loss among these subsets by examining the gene expression profiles in Ring1A/B-dKO ESCs ( Figure 1C; Figure S2B ) . We found significant de-repression ( p<0 . 001 ) of TP , DP and SP genes but no significant changes in H3K27me3-negative genes . It is worth noting that the degree of de-repression of the TP genes was significantly higher than that of the DP and SP genes ( Figure 1C ) . Gene ontology ( GO ) based analyses confirmed that TP genes are most significantly enriched for functions in transcription and/or development ( Figure S5 ) . Of note , Cdx2 and Gata6 , which are known to be repressed by Oct3/4 and Nanog [29] , [30] , are occupied by H2AK119u1 , suggesting that H2AK119u1 might be involved in maintaining ESC properties by suppressing differentiation of ESCs . Above data suggest a potential importance of H2AK119u1 for repression of key developmental regulators which is required to maintain the undifferentiated status of ESCs . This however does not necessarily prove the importance of the E3 ligase activity and H2AK119u1 per se for the repression because H2A ubiquitination independent functions of PRC1 in chromatin compaction and gene silencing both in vitro [24] , and in vivo [23] has been reported in previous studies . To investigate this question , we expressed mutant Ring1B proteins that are defective in the interaction with the E2 component in Ring1A/B-dKO ESCs . In this experimental setup , we have first stably expressed exogenous WT or mutant Ring1B in Ring1A−/−;Ring1Bfl/fl;R26::CreERT2 ESCs and then endogenous Ring1B was depleted by OHT treatment ( Figure 2A ) . Similar experiments have been described previously [23] but have made use of Ring1B single KO ESCs , and therefore could not exclude the contribution of low levels of Ring1A ( Figure S6A ) and associated H2AK119u1 that occur in ESCs [5] , [17] , [20] , [25] . We thus tested the role of Ring1A to mediate H2AK119u1 and repression of Polycomb targets in ESCs . We first compared global H2AK119u1 levels in Ring1B-KO ESCs with Ring1A/B-dKO and found significant amount of H2AK119u1 remained in Ring1B-KO ( Figure S6B ) . Consistently , the expression of exogenous Ring1A obviously restored global H2AK119u1 level , ESC identity , and repression of TP genes in Ring1A/B-dKO ESCs ( Figure S6B–E ) . We then performed ChIP-chip analysis to compare Ring1A distribution with Ring1B and found that Ring1A and Ring1B significantly shared target genes ( Figure S6F , G ) . ChIP-qPCR analysis further confirmed Ring1A binding at promoter regions of representative TP genes such as Hoxd11 and Zic1 in the absence of Ring1B ( Figure S6H ) . Importantly , binding of other PRC1 components such as Mel18 was concomitantly restored by the expression of exogenous Ring1A in Ring1A/B-dKO ESCs ( Figure S6H ) . Therefore , Ring1A was shown to substitute for Ring1B functions in mediating H2AK119u1 and target gene repression . These results sufficiently justify the use of Ring1A/B-dKO ESCs instead of Ring1B-KO in following experiments . We made use of the previously characterized I53S and I53A mutations located at the E2 UbcH5c binding surface that have been shown to affect the E3 activity of Ring1B both in vitro and in vivo [17] , [23] , [31] . We introduced expression vectors for flag-tagged wild-type ( WT ) or mutant Ring1B [Ring1B ( I53S ) or ( I53A ) ] into Ring1A−/−;Ring1Bfl/fl;R26::CreERT2 ESCs ( Figure 2A ) and established stable transfectants that expressed exogenous Ring1B at similar level to the endogenous protein ( Figure 2B ) . Expression of WT Ring1B restored global H2AK119u1 levels in Ring1A/B-dKO cells whereas Ring1B ( I53S ) and Ring1B ( I53A ) did not ( Figure 2B ) . We went on to check whether the levels and target binding of PRC1 could be appropriately recapitulated by exogenous wild-type or mutant Ring1B in the transfectants . Levels of other PRC1 proteins were depleted in the absence of Ring1B , presumably because complex formation stabilizes individual components [9] , [26] . We found that Mel18 was clearly detectable and formed complexes with Ring1B ( I53S ) , Ring1B ( I53A ) and wild-type Ring1B in the absence of endogenous Ring1 proteins in each transfectant ( Figure 2C ) . We also confirmed that levels of Cbx2 and Phc1 were restored in these transfectants ( data not shown ) . We next assessed the association of exogenous Ring1B with target genes in the transfectants . We used ChIP and subsequent quantitative PCR ( qPCR ) analysis and observed binding of Ring1B I53S or I53A to target loci . Local H2AK119u1 deposition was undetectable , confirming the impaired E3 ligase activity of the Ring1B mutant proteins ( Figure 2D ) . We also found that Mel18 binding to these targets was considerably restored by the expression of Ring1B ( I53S ) or Ring1B ( I53A ) . Finally , we tested whether condensation of Hoxb cluster could be recapitulated in the transfectants by using 3D FISH analysis with probes for Hoxb1 and Hoxb13 . Consistent with a previous report using Ring1B-KO ESCs , we found that Hoxb1 and Hoxb13 were considerably separated in Ring1A/B-dKO ESCs compared to Ring1A−/− cells ( Figure 2E ) [23] and that condensation of the Hoxb cluster was significantly restored by the expression of Ring1B ( I53S ) or Ring1B ( I53A ) . Taken together , the expression and target binding of PRC1 were sufficiently recapitulated in Ring1A/B-dKO ESCs that express catalytically inactive Ring1B . We thus concluded that these transfectants were well suited to address the role of Ring1B E3 activity in the maintenance and repression of ESC Polycomb targets . The above results also imply that E3 activity of Ring1B and H2AK119u1 are dispensable for PRC1 target binding . We next tested the phenotypes of the transfectants after deletion of endogenous Ring1B . We observed that the expression of either Ring1B ( I53S ) or Ring1B ( I53A ) was not sufficient to maintain ESCs in an undifferentiated state ( Figure 3A ) . We obtained similar results in the presence of three inhibitors ( 3i ) that target FGF receptor , MEK , and GSK3 ( data not shown ) [32] . This implies that the E3 activity of Ring1B is required to maintain ESC identity in the absence of Ring1A . Consistently , Ring1B ( I53S ) failed to restore repression of differentiation markers ( Kdr , Gata6 , Hnf4a and Cdx2 ) and expression of undifferentiation markers ( Pou5f1 , Sox2 and Nanog ) in Ring1A/B-dKO ESCs while WT Ring1B or Ring1A obviously restored ( Figure S7A ) . We went on to examine differentiation ability of the respective ESC lines by forming embryoid bodies . We found that the progressive changes in expression of the marker genes upon induction of differentiation were considerably affected in Ring1A/B-dKO ESCs compared to wild-type or Ring1A−/− ESCs ( Figure S7B ) . These changes were restored by WT Ring1B but not by Ring1B ( I53S ) . Together , Ring1B catalytic activity is required for maintenance and differentiation of ESCs . We then examined the expression of H2AK119u1+ genes in these transfectants by using expression microarrays . In the mock transfectant , we observed that H2AK119u1+ genes were significantly de-repressed by depletion of Ring1 proteins whereas expression of H2AK119u1− genes was virtually unchanged ( Figure 3B ) . De-repression of H2AK119u1+ genes in the Ring1A/B-dKO was mostly restored by the expression of WT Ring1B but only partially by Ring1B ( I53S ) and Ring1B ( I53A ) ( Figure 3B; Figure S8 ) . Moreover , the levels of restoration by Ring1B mutants were variable among target genes . To confirm the microarray results , we examined the expression of H2AK119u1 targets , Hoxa9 , Hoxb13 , Hoxd11 , Zic1 , and Pax3 , by quantitative RT-PCR . These genes were de-repressed in Ring1A/B-dKO compared to OHT-untreated control cells ( Figure 3C ) . WT Ring1B was shown to fully restore the repression of these genes . Ring1B ( I53S ) and Ring1B ( I53A ) could slightly restore the repression of Hoxa9 , Hoxb13 and Hoxd11 , but almost failed to repress Zic1 and Pax3 ( Figure 3C ) . Therefore , the E3 activity of Ring1B is required for efficient repression of its target genes . Our results also suggest that some genes , e . g . , Zic1 and Pax3 are more dependent on the E3 activity than others , notably the Hox cluster genes such as Hoxd11 . The above experiments strongly suggest that repression of developmental regulators in ESCs is attributable to PRC1 mediated H2AK119u1 . Previous studies report that Ring1B regulates local H3K4me3 deposition and loading of RNA polymerase II ( RNAP ) in ESCs [5] , [19] , and that H2AK119u1 has a role to suppress MLL-mediated methylation of H3 lysine 4 ( H3K4 ) and transcriptional initiation from nucleosomal templates [28] . We , therefore , examined whether the catalytic activity of Ring1B is involved in suppressing H3K4 methylation and RNAP loading at target gene loci . Consistent with the previous reports , we found that local levels of trimethylated H3K4 ( H3K4me3 ) and RNAP loading were considerably up-regulated at target gene promoters in Ring1A/B-dKO ESCs , which could be repressed by expression of WT Ring1B in these cells ( Figure 3D and 3E ) . In contrast , Ring1B ( I53S ) and Ring1B ( I53A ) failed to suppress local increases of H3K4me3 and RNAP levels . Therefore , the catalytic activity of Ring1B is required to repress H3K4me3 and RNAP loading . Consistent with these observations , the profound reduction in local H3K27me3 levels in Ring1A/B-dKO ESCs could not be restored by Ring1B ( I53S ) or Ring1B ( I53A ) ( Figure S9 ) . This may also suggest the contribution of Ring1B catalytic activity to maintain repressive chromatin . Collectively , our results demonstrate that Ring1B-dependent H2AK119u1 facilitates transcriptional repression of PRC1 target genes and thereby enables the maintenance of ESC identity .
In the present study , we present genome-wide H2AK119u1 maps in ESCs and identify a group of genes at which H2AK119u1 is deposited in a Ring1-dependent manner . These genes are a distinctive subset of genes with H3K27me3 enrichment and we suggest that these are the central targets of Polycomb silencing to maintain ESC identity . By using mutant versions of Ring1B , which can not bind to E2 components , we demonstrate the role of H2AK119u1 to facilitate the repression of these target genes . We propose that H2AK119u1 contributes to capacitate Polycomb-mediated repression in a reversible manner because recognition and de-ubiquitination of H2AK119u1 have been shown to be linked with transcriptional activation [27] , [28] , [33] . This conclusion is different to a recent study which suggested that the catalytic mutant Ring1B could restore repression in Ring1B mutant ES cells [23] . A key difference in that study and our analyses shown here is that we assessed the function of catalytically inactive Ring1B in a background that is null for both Ring1B and the closely related homologue Ring1A . Ring1A potentially complements loss of Ring1B in ESCs , despite the fact that the expression level of Ring1A is relatively low compared to Ring1B ( Figure S6 ) [5] , [20] . Our results are concordant with those of Eskeland et al . 2010 which reports that the ability of Ring1B to mediate the condensation of the Hoxb cluster is not dependent on its histone ubiquitination activity . In addition , in our study we have observed that the E3 activity of Ring1B contributes to the repression of Hox genes to a lesser extent than to Zic1 and Pax3 genes ( Figure 3C ) . Based on these evidences , we propose that H2AK119u1-dependent repression is likely complemented by other PRC1-mediated mechanisms such as chromatin compaction [23] . The fact that H2AK119u1 independent repression is more prevalent at Hox loci compared to other Polycomb target genes may suggest that it is more effective when target loci are closely juxtaposed in cis . We indeed found a slight but a significant restoration of repression of H2AK119u1+ genes that are closely juxtaposed each other ( <50 kb ) by expression of mutant Ring1B in Ring1A/B-dKO , but this is not the case for H2AK119u1+ genes that are separated by ≥50 kb genomic regions ( Figure S10 ) . However , this tendency is not statistically significant once we excluded Hox cluster genes . Further studies are needed to elucidate the molecular nature for H2AK119u1-independent mechanisms . Overall , our findings show that PRC1 mediates gene repression by combining multiple and different effector mechanisms , of which H2A ubiquitination is a major contributor ( Figure 4 ) . Such diverse PRC1 effector mechanisms might be required to make PRC1-mediated gene repression both flexible and robust . How H2A ubiquitination contributes to repress target gene transcription also awaits future studies , although mechanisms that antagonize against H2A ubiquitination have already been proposed [27] .
Ring1Bfl/fl;Rosa26::CreERT2 , Ring1A−/−;Ring1Bfl/fl;Rosa26::CreERT2 , and Eed-KO ESCs were described previously [5] , [19] , [20] , [26] , [34] . The ESCs were cultured in DMEM with 20% fetal bovine serum , MEM nonessential amino acids ( Invitrogen ) , sodium pyruvate ( Invitrogen ) , L-glutamine ( Invitrogen ) , 2-mercaptoethanol ( Sigma ) , and ESGRO ( Chemicon ) either on irradiated MEF as feeder layers or directly on gelatin-coated surfaces . 3xFlag-tagged wild-type Ring1A , wild-type Ring1B , mutated Ring1B ( I53A [17] , [23] and I53S [31] ) cDNAs were subcloned into the expression vector pCAG-IRES-Puro ( a kind gift from Dr . Hitoshi Niwa in RIKEN CDB in Japan ) . The following antibodies were used: Ring1B ( clone #3 ) [35] , Ring1A [36] , Phc1 [37] , Mel18 ( Santa Cruz; sc-10744 ) , Cbx2 [36] , H3K27me3 ( Millipore; 07-449 ) , H3K4me3 ( Millipore;07-473 ) , H2AK119u1 ( E6C5; Millipore; 05-678; for ChIP ) , H2AK119u1 ( Rabbit polyclonal; for ChIP ) [28] , H2AK119u1 ( Rabbit polyclonal; Cell Signaling Technology; #8240; for western blot ) , H2A ( Abcam; ab18255 ) , RNAP ( 8WG16; Millipore; 05-952 ) , Lamin B ( Santa Cruz; sc-6216 ) , mouse IgM ( Millipore; 12-488 ) , and Flag-tag ( M2; Sigma; F1804 ) . Ring1A−/−; Ring1Bfl/fl; Rosa26::CreERT2 ESCs were stably transfected with tagged wild-type Ring1A , wild-type Ring1B , or mutated Ring1B . To establish stable transfectants , ESCs were electroporated ( 0 . 8 kV , 3 µF ) with the respective expression vector and then selected for resistance to puromycin ( 1 µg/ml ) . Cells expressing each of tagged constructs were suspended in IP buffer [10 mM Tris-HCl ( pH8 . 0 ) , 1 mM EDTA , 140 mM NaCl , 0 . 4% NP-40 , and 0 . 5 mM PMSF] and sonicated for several seconds . After centrifugation , the supernatant was collected , precleared with protein G Sepharose for 30 min at 4°C , and then incubated with anti-Flag antibody ( M2 ) for 120 min at 4°C . The immune complexes were captured by protein G Sepharose for 60 min at 4°C . The Sepharose-bound proteins were washed with IP buffer , eluted in SDS sample buffer under reducing condition , separated on SDS-PAGE gels , and subjected to western blot analysis . Quantitative real-time PCR was carried out with SYBR Green method and amplifications were detected with Mx3005P ( Stratagene , La Jolla , CA , USA ) . The sequences of primers used in this study are shown in Text S1 . ESCs were treated with 1% formaldehyde/PBS for 10 min at room temperature . Cells were washed with PBS , collected and resuspended in swelling buffer [20 mM Hepes ( pH 7 . 8 ) , 1 . 5 mM MgCl2 , 10 mM KCl , 0 . 1% NP-40 , and 1 mM DTT] by pipetting and then kept on ice for 10 min . After Dounce homogenizing 10–20 times , the cells were centrifuged and then the pellets were resuspended in RIPA buffer [20 mM Tis-HCl ( pH 8 . 0 ) , 1 mM EDTA , 140 mM NaCl , 1% Triton X-100 , 0 . 1% SDS , and 0 . 1% deoxycholic acid] containing protease inhibitors and sonicated into fragments with an average length of 0 . 3–0 . 5 kb . After centrifugation , the supernatants were subjected to IP with specific antibodies as previously described [19] , [38] . For H2AK119u1-ChIP , pre-cleared chromatin ( 400 µg ) was incubated with 50 µl of E6C5 antibody ( overnight , 4°C ) and then the chromatin-1st antibody complexes were immunoprecipitated with 2nd antibody ( rabbit anti-mouse IgM ) - preconjugated protein A dynabeads ( Invitrogen ) . Purified immunoprecipitated and input DNA was quantified by real-time PCR , and , if necessary , was subjected to the linear amplification for ChIP-chip analysis . ChIP-on-chip analysis was carried out using the Mouse Promoter ChIP-on-chip Microarray Set ( G4490A , Agilent , Palo Alto , Calif . , USA ) . ESCs were subjected to ChIP assay using specific antibodies as described in the previous section . Purified immunoprecipitated and input DNA was subjected to T7 RNA polymerase-based amplification as described previously [39] . Labeling , hybridization and washing were carried out according to the Agilent mammalian ChIP-on-chip protocol ( ver . 9 . 0 ) . Scanned images were quantified with Agilent Feature Extraction software under standard conditions . All of experiments were performed by using at least two biological replicates . The obtained data were analyzed as described in Text S1 . Total RNA was extracted using the Trizol reagent ( Invitrogen , Carlsbad , CA , USA ) and purified with Qiagen RNeasy separation columns ( Qiagen , Hilden , Germany ) . First strand cDNA was synthesized and hybridized to Affymetrix GeneChip Mouse Genome 430 2 . 0 arrays ( Affymetrix , Santa Clara , CA , USA ) to assess and compare the overall gene expression profiles . The obtained data were analyzed as described in Text S1 . 3D-DNA-FISH with spatial preservation of chromatin architecture was performed as described previously [40] . Experimental details are described in Text S1 . ChIP-chip and microarray data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus and is accessible through GEO Series accession number GSE38650 . | Polycomb-group ( PcG ) proteins play essential roles in the epigenetic regulation of gene expression during development . PcG proteins form two distinct multimeric complexes , PRC1 and PRC2 . In the widely accepted hierarchical model , PRC2 is recruited to specific genomic locations and catalyzes trimethylation of H3 lysine 27 ( H3K27me3 ) , thereby creating binding sites for PRC1 , which then catalyzes mono-ubiquitination of histone H2A ( H2AK119u1 ) . Recently , PRC1 has been shown to be able to compact chromatin structure at target loci independently of its histone ubiquitination activity . Therefore , the role of H2AK119u1 still remains unclear . To gain insight into this issue , we used ChIP-on-chip analysis to map H2AK119u1 genome-wide in mouse ES cells ( ESCs ) . The data demonstrate that H2AK119u1 occupies a distinctive subset of genes with H3K27me3 enrichment . These genes are the central targets of Polycomb silencing to maintain ESC identity . We further show that the H2A ubiquitination activity of PRC1 is dispensable for its target binding and its activity to compact chromatin at Hox loci , but is indispensable for efficient repression of target genes and therefore ESC maintenance . We propose that multiple effector mechanisms including H2A ubiquitination and chromatin compaction combine to mediate PRC1-dependent repression of developmental genes to maintain the identity of ESCs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"gene",
"regulation",
"cell",
"differentiation",
"histone",
"modification",
"developmental",
"biology",
"genome",
"analysis",
"tools",
"stem",
"cells",
"epigenetics",
"molecular",
"genetics",
"chromatin",
"chromosome",
"biology",
"... | 2012 | Histone H2A Mono-Ubiquitination Is a Crucial Step to Mediate PRC1-Dependent Repression of Developmental Genes to Maintain ES Cell Identity |
Leishmania is a protozoan parasite that alternates its life cycle between the sand-fly vector and the mammalian host . This alternation involves environmental changes and leads the parasite to dynamic modifications in morphology , metabolism , cellular signaling and regulation of gene expression to allow for a rapid adaptation to new conditions . The L-arginine pathway in L . amazonensis is important during the parasite life cycle and interferes in the establishment and maintenance of the infection in mammalian macrophages . Host arginase is an immune-regulatory enzyme that can reduce the production of nitric oxide by activated macrophages , directing the availability of L-arginine to the polyamine pathway , resulting in parasite replication . In this work , we performed transcriptional profiling to identify differentially expressed genes in L . amazonensis wild-type ( La-WT ) versus L . amazonensis arginase knockout ( La-arg- ) promastigotes and axenic amastigotes . A total of 8253 transcripts were identified in La-WT and La-arg- promastigotes and axenic amastigotes , about 60% of them codifying hypothetical proteins and 443 novel transcripts , which did not match any previously annotated genes . Our RNA-seq data revealed that 85% of genes were constitutively expressed . The comparison of transcriptome and metabolome data showed lower levels of arginase and higher levels of glutamate-5-kinase in La-WT axenic amastigotes compared to promastigotes . The absence of arginase activity in promastigotes increased the levels of pyrroline 5-carboxylate reductase , but decreased the levels of arginosuccinate synthase , pyrroline 5-carboxylate dehydrogenase , acetylornithine deacetylase and spermidine synthase transcripts levels . These observations can explain previous metabolomic data pointing to the increase of L-arginine , citrulline and L-glutamate and reduction of aspartate , proline , ornithine and putrescine . Altogether , these results indicate that arginase activity is important in Leishmania gene expression modulation during differentiation and adaptation to environmental changes . Here , we confirmed this hypothesis with the identification of differential gene expression of the enzymes involved in biosynthesis of amino acids , arginine and proline metabolism and arginine biosynthesis . All data provided information about the transcriptomic profiling and the expression levels of La-WT and La-arg- promastigotes and axenic amastigotes . These findings revealed the importance of arginase in parasite survival and differentiation , and indicated the existence of a coordinated response in the absence of arginase activity related to arginine and polyamine pathways .
Leishmania is a protozoan parasite that causes widespread human disease known as leishmaniases , characterized by cutaneous , mucosal or visceral manifestations . Leishmania alternates its life cycle between the sand-fly vector ( promastigote form ) and the mammalian host ( amastigotes form ) [1] . This alternation involves environmental changes and submits the parasite to dynamic modifications in morphology , metabolism , cellular signaling and regulation of gene expression to allow for a rapid adaptation to new conditions . The parasite has also developed resistance mechanisms to evade sand-fly digestive enzymes and the host innate immune response , such as the mammalian complement system and macrophage defense mechanisms involving nitric oxide ( NO ) . NO is produced by nitric oxide synthase 2 ( NOS2 ) using the amino acid L-arginine as substrate [2 , 3] . On the other hand , arginase is an immune-regulatory enzyme that can reduce NO production by activated macrophages , limiting the availability of L-arginine to NOS2 , supporting Leishmania resistance to host defense mechanisms . Arginase uses L-arginine to produce urea and ornithine , a precursor of the polyamine pathway [4] . The success of the Leishmania infection depends on the parasite ability to subvert the host defense mechanisms [3 , 5] . Leishmania also expresses arginase , which supplies the metabolic precursors for parasite replication , an essential step for the establishment of the infection [4 , 6] . Our research group has been studying the role of arginase in L . amazonensis during the parasite life cycle and its role in the establishment and maintenance of the infection in mammalian macrophages [7–9] . Transcriptional profiling has been used in expression studies of several model organisms , including Leishmania [10–13] . Holzer et al . ( 2006 ) used microarray analysis to determine that 3 . 5% of the genes were differentially expressed between promastigotes and lesion-derived amastigotes of L . mexicana , and 0 . 2% were differentially expressed between promastigotes and axenic amastigotes . The reduced number of regulated genes was a consequence of an increase in the magnitude of the transcript levels in cells under axenic conditions [14] . Leifso et al . ( 2007 ) also demonstrated differential gene expression between promastigote and lesion-derived amastigote forms of L . major [15] . These data indicated that the Leishmania genome is mostly constitutively expressed during the parasite life cycle , but there are still some genes that are differentially expressed to adapt to different environmental changes [14–16] . In addition , Goldmann et al . ( 2007 ) demonstrated with transcriptome analysis that arginase has an important role in the establishment of infection with Streptococcus pyogenes . Arginase type II was up-regulated in the infection of macrophages with S . pyogenes after 1 , 4 and 16 h . However , NOS2 did not show differential gene expression . The same profile was observed in macrophages stimulated with γ-interferon and lipopolysaccharide [17] . In this work , through RNA-seq of La-WT and La-arg- promastigotes and axenic amastigotes , we identified 8253 transcripts , from which 60% encoding hypothetical proteins and 443 novel transcripts that did not match any previously annotated gene . The transcriptional profiling revealed that 85% of the genes were constitutively expressed . Among the 15% ( 1268 genes ) that were DE , we identified genes up- and down-regulated . Interestingly , we showed 100 genes differentially expressed in La-WT promastigotes and 908 genes differentially expressed in La-arg- promastigotes . Additionally , we identified 183 genes differentially expressed in La-WT axenic amastigotes and only 34 genes differentially expressed in La-arg- axenic amastigotes . In summary , our results showed that L . amazonensis could modulate gene expression with differential regulation between promastigote and axenic amastigotes , indicating that this organism may represent an alternative paradigm for eukaryotic differentiation with minimal contributions from changes in mRNA abundance . The transcriptional profiling also revealed differential gene expression in the development of the Leishmania life cycle and the existence of a coordinated response in the absence of arginase activity , providing additional insights into how Leishmania is able to modulate its biological functions to survive during environmental changes .
Leishmania ( Leishmania ) amazonensis ( MHOM/BR/1973/M2269 ) , a strain of our laboratory collection at the Institute of Bioscience , and L . amazonensis arginase knockout ( La-arg- ) [8] promastigotes were grown at 25°C in M199 medium , pH 7 . 0 , supplemented with L-glutamine , 10% heat-inactivated fetal bovine serum , 0 . 25% hemin , 40 mM NaHCO3 , 100 μM adenine , 40 mM HEPES , 100 U/mL penicillin and 100 μg/mL streptomycin . Axenic amastigotes of La-WT and La-arg- were grown in M199 medium supplemented , as described above at 34°C , pH 5 . 5 . For the La-arg- cultures , hygromycin ( 30 μg/mL ) , puromycin ( 30 μg/mL ) and putrescine ( 50 μM ) were added . Bone marrow derived-macrophages ( BMDM ) were collected from the femur of female BALB/c mice ( 6–8 weeks ) from the Animal Center of the Institute of Bioscience of the University of Sao Paulo . The femurs were washed with cold PBS and the cells were collected at 500 x g for 10 min at 4°C . The lysis of erythrocytes was performed with NH4Cl ( 145 mM ) and Tris-base ( 200 mM ) pH 7 . 0 and incubated on ice for 20 min . After lysis , the cells were washed with cold PBS , collected at 500 x g for 10 min at 4°C and incubated in RPMI 1640 medium supplemented with penicillin ( 100 U/mL ) , streptomycin ( 100 μg/ml ) , 2-mercaptoethanol ( 50 μM ) , L-glutamine ( 2 mM ) , sodium pyruvate ( 1 mM ) , fetal bovine serum 10% and L929 conditioned medium ( 15% ) , as macrophage stimulating factor source . The cells were cultivated for 7 days at 34°C and 5% CO2 . After differentiation , cellular viability was evaluated with Trypan blue staining 1:1 , and cells were counted in a Neubauer chamber . Approximately 1x106 BMDM were incubated on sterile 13-mm coverslips in 24-well plates overnight at 34°C and 5% CO2 to adhere to the coverslips . Non-adherent cells were removed by PBS washing , and the infection was performed with La-WT or La-arg- axenic amastigotes ( MOI 5:1 ) . After 4 h of infection , the cultures were washed with PBS and maintained in culture for 24 , 48 and 72 h . Non-infected macrophages maintained in culture at the same conditions were used as control . The infections were evaluated by determining the percentage of infection after counting 200 Panoptic-stained ( Laborclin , Parana , Brazil ) macrophages . The infection index was determined by multiplying the percentage of infected macrophages by the mean number of parasites per infected cell [18 , 19] . Statistical analyses were performed using the t-test . Total RNA from 3 independent biological replicates was isolated from La-WT and La-arg- promastigotes and axenic amastigotes using TRIzol reagent ( Life Technologies , Carlsbad , CA , USA ) , according to the manufacturer’s instructions . RNA samples were treated with DNase I ( Thermo Scientific , Lithuania , EU ) , and the RNA concentration was determined using a spectrophotometer at A260/A280 ( Nanodrop ND1000 , Thermo Scientific , USA ) . In addition , the RNA integrity was evaluated using an Agilent 2100 Bioanalyzer and Pico Agilent RNA 6000 kit ( Agilent Technologies , Santa Clara , CA , USA ) , according to the manufacturer’s instructions . rRNA depletion was performed by poly ( A ) magnetic beads capture protocol , using Strand-specific TrueSeq RNA-seq Library Prep ( Illumina ) , according to manufacturer´s instruction . Library preparations were performed using Strand-specific TrueSeq RNA-seq Library Prep ( Illumina ) , according to the manufacturer’s instructions . Paired-end reads ( 125 bp ) were obtained using the Illumina HiSeq 2000 platform at the Norwegian Sequencing Centre at the University of Oslo . Trimmomatic was used to remove the Illumina adapter sequences [20] . The quality of the produced data was analyzed using FastQC by Phred quality score [21] . Reads with Phred quality scores lower than 20 were discarded . Reads were aligned to the L . mexicana ( MHOMGT2001U1103 ) genomic data obtained from TriTrypDB version 29 ( www . tritrypdb . org ) using TopHat ( -G option ) [22 , 23] . Maximum 2 mismatches were allowed . Thereafter , the expression level of the assembled transcriptome and abundance estimation were performed using Cufflinks [24] . The abundance of transcripts was calculated as the Fragments Per Kilobase of transcript per Million mapped reads ( FPKM ) , which reflects the abundance of a transcript in the sample by normalization of the RNA length and the total read number [25] . The gene expression level values were calculated from the transcript counts . Differentially expressed gene analysis was performed on four comparisons pairs ( La-WT promastigotes vs . La-arg- promastigotes; La-WT axenic amastigotes vs . La-arg- axenic amastigotes , La-WT promastigotes vs . La-WT axenic amastigotes; La-arg- promastigotes vs . La-arg- axenic amastigotes ) . Genes with zero FPKM were excluded ( excepted the arginase gene ( LmxM . 34 . 1480 ) ) . Transcripts that did not match any previous annotated gene were considered novel , and they were identified using Cufflinks with -g option . Statistical significance of DE genes data was determined using independent t-test and fold change in which the null hypothesis was that no difference exists among groups . False discovery rate ( FDR ) was controlled by adjusting p value using Benjamini-Hochberg algorithm [26] . Functional annotation was performed using GO ( Gene Ontology ) and the Kyoto Encyclopedia of Genes and Genomes ( KEGG ) . All analyses were performed by Macrogen ( www . macrogen . com ) . Reverse transcription was performed using 2 μg of total RNA as a template , reverse transcriptase and random primers ( Revertaid H minus Reverse Transcriptase kit , Thermo-Scientific , Canada ) , according to the manufacturer’s instructions . Equal amounts of cDNA were assessed in triplicate in a total volume of 25 μL containing Power SYBR Green qPCR Master Mix ( Life Technologies , Warrington , UK ) and the following primers ( 20 μM ) : GAPDH_F 5´-TCAAGGTCGGTATCAACGGC-3´ , GAPDH_R 5´-TGCACCGTGTCGTACTTCAT-3´ , arginase F 5´-TCCTGCACGACCTGAACATC-3´ , arginase R 5´-CGCCATGGACACCACCTT-3´ , glutamate 5-kinase F 5´-AGCTGGTTTTTGGCGACAAC-3´ , glutamate 5-kinase R 5´-CGTCGATGTCGCTGAGAATG-3´ , pyrroline 5-carboxylase dehydrogenase F 5´-ACGGTGTTTGTGTATGACGACAGT-3´ , pyrroline 5-carboxylase dehydrogenase R 5´-ACCGGTCAGGCCGTACTTC-3´ , spermidine synthase F 5´-GCAACCAGGGCGAGTCTATCT-3´ , spermidine synthase R 5´-TGACCGTGGAAAAGCCAATAT-3´ , amastin-like F 5´-GGAGCGCTACTTCAGCTATGGA-3´ , amastin-like R 5´-CGGATCATCAATAAGACGATGTTG-3´ , amastin-like F 5´-CGGCTGCCTTTTGCTGTACT-3´ and amastin-like R 5´-CAGACAACGCAAGCTGTGACA-3´ . The mixture was incubated at 94°C for 5 min , followed by 40 cycles at 94°C for 30 s , 60°C for 30 s and 72°C for 30 s . A negative control in the absence of reverse transcriptase was included in RT-qPCR assays to detect DNA contamination in RNA samples . The copy number of the target and reference genes were quantified in three biological samples , considering the molar mass concentration , according to a standard curve generated from a ten-fold serial dilution of a quantified PCR product . The normalized target/gapdh ratio of the molecules absolute number of each target was used as a parameter of the expression . Reactions were carried out using PikoReal 96 RealTime PCR System ( Thermo Scientfic , Finland ) . Analyses were performed using PikoReal Software 2 . 2 ( Thermo Scientific , Finland ) . The experimental protocols for the animals were approved by the Animal Care and Use Committee from the Institute of Bioscience of the University of Sao Paulo ( CEUA 196/2014 ) . This study was carried out in strict accordance with the recommendations in the guide and policies for the care and use of laboratory animals of the São Paulo State ( Lei Estadual 11 . 977 , de 25/08/2005 ) and Brazil government ( Lei Federal 11 . 794 , de 08/10/2008 ) .
La-WT and La-arg- promastigotes in the stationary growth phase were submitted to differentiation in medium , pH 5 . 5 at 34°C , for 48 h . The growth curve of the differentiated amastigotes of both La-WT and La-arg- reached stationary growth phase after 10 days of incubation ( S1A Fig ) . Axenic amastigotes were also induced to differentiate back to promastigotes incubating the parasites at pH 7 . 0 and 26°C that led to differentiation after 48 h of incubation . Moreover , axenic amastigotes infectivity was evaluated . BMDM from BALB/c mice were infected with La-WT and La-arg- ( MOI 5:1 ) , and the infection index was analyzed at 24 , 48 and 72 h post infection . According to S1B Fig , both La-WT and La-arg- axenic amastigotes were able to infect and establish the infection . However , the infection index for La-arg- was lower than for La-WT , corroborating the previously determined infection index for La-arg- promastigotes [8] . Once axenic amastigotes infectivity was confirmed , the total RNA from La-WT and La-arg- promastigotes and axenic amastigotes was extracted and submitted to RNA-seq , as described in Methods . Transcriptomic analyses were performed using 3 independent biological replicates from each La-WT and La-arg- promastigotes and axenic amastigotes after Illumina HiSeq2000 sequencing that generated million sequence reads ( 125 bp ) ( S1 Table ) . Sequencing data are available on the NCBI BioProject under accession number PRJNA380128 and Sequence Read Archive ( SRA ) under accession number SRX2661998 and SRX2661999 . RNA-seq data were aligned to the L . mexicana genome [27] . After initial assembling , 8253 transcripts and 443 novel transcripts were identified with genome coverage around 90% . And 60% of the transcripts corresponding to hypothetical proteins as listed in S2 Table . The novel transcripts were those that did not correspond to any previous annotated gene , even in L . mexicana genome , used in the comparisons since L . amazonensis is not completely annotated . According to Fig 1 we demonstrated the arginase transcript ( LmxM . 34 . 1480 ) assembling with genome coverage in La-WT promastigotes and axenic amastigotes . As expected no transcripts were detected in the knockout lines ( La-arg- promastigotes and axenic amastigotes ) . RNA-seq has been described as an accurate method for quantifying transcript levels [10–12 , 28 , 29] . Our RNA-seq data revealed that 85% of the genes were constitutively expressed , comparing the gene expression profiles of La-WT and La-arg- promastigotes and axenic amastigotes . However , among 15% ( 1268 genes ) DE genes , we identified a vast number of genes differentially expressed . Of the total 378 and 357 DE genes in La-WT promastigotes and axenic amastigotes , 100 and 183 genes were non-common for each line , respectively . Of the total 908 and 62 DE genes in La-arg- promastigotes and axenic amastigotes , 554 genes and 34 genes were non-common for each line , respectively ( Fig 2 ) . A direct overlap revealed only 2 transcripts mutually expressed among all samples ( Fig 2 ) . The analyses of the DE genes , limited to those presented fold change ≥ 2 and p ˂ 0 . 05 , revealed 195 genes up-regulated and 183 genes down-regulated in the comparison of La-WT promastigote vs . La-arg- promastigote , suggesting a significant amount of DE genes which regulation depends on arginase activity . On the other hand , in the comparison of La-arg- axenic amastigotes vs . La-WT axenic amastigotes , only 37 genes were up-regulated and 25 genes were down-regulated . In addition , in the comparison of La-WT axenic amastigotes vs . La-WT promastigotes we observed 208 up-regulated and 149 down-regulated genes , indicating a significant amount of DE genes during La-WT differentiation . The comparison of La-arg- axenic amastigotes vs . La-arg- promastigotes led to larger number of DE genes ( 452 up-regulated and 456 down-regulated genes ) what could be explained considering the two variables in this comparison , the absence of arginase activity and the life cycle stage ( Fig 3 ) . The additional characterization of axenic amastigotes transcripts revealed up-regulation of the following amastins: amastin ( LmxM . 30 . 0451 and LmxM . 30 . 0452 ) and amastins-like ( LmxM . 08 . 0750 , LmxM . 08 . 0760 , LmxM . 08 . 0770 , LmxM . 08 . 0800 , LmxM . 08 . 0850 , LmxM . 33 . 0960 , LmxM . 33 . 0961 , LmxM . 33 . 1560 and LmxM . 33 . 1920 ) ( S2 Fig ) . Based on the DE genes analyzed , we generated volcano plots showing the distribution of transcripts by comparing the fold change in the expression ( log2 ) of each group with the corresponding adjusted p value ( -log10 ) ( S3 Fig ) . We further analyzed the volume plot ( S4 Fig ) identifying the top 5 transcripts that showed higher expression difference compared to the control according to expression volume ( Table 1 ) . Comparing the expression profile of La-WT vs . La-arg- promastigotes , we identified the following up-regulated transcripts: a conserved hypothetical protein ( LmxM . 15 . 1520 ) and a non-coding RNA ( LmxM . 23 . ncRNA rfamscan: 218578-218718-1 ) ; and the following down-regulated transcripts: a putative nucleolar RNA-binding protein ( LmxM . 07 . 0990 ) and two histone H4 proteins ( LmxM . 06 . 0010 and LmxM . 15 . 0010 ) . Comparing the expression of La-WT vs . La-arg- axenic amastigotes , we identified the following up-regulated transcripts: a putative dipeptidyl-peptidase III ( LmxM . 05 . 0960 ) , the protein disulfide isomerase ( LmxM . 06 . 1050 ) and a conserved hypothetical protein ( LmxM . 08 . 0540 ) ; and the following down-regulated transcripts: a conserved hypothetical protein ( LmxM . 17 . 0890 ) and a putative tryparedoxin 1 protein ( LmxM . 08_29 . 1160 ) . The comparison of the expression profile of La-WT promastigotes vs . axenic amastigotes led to the identification of the following up-regulated transcripts: the tryparedoxin peroxidase ( LmxM . 15 . 1160 ) and a putative ATP-dependent RNA helicase ( LmxM . 33 . 2050 ) ; and the following down-regulated transcripts: a putative histone H3 ( LmxM . 10 . 0970 ) , the histone H4 ( LmxM . 15 . 0010 ) and an unspecific product ( LmxM . 13 . 0290partial ) . Finally , the comparison of the expression profile of La-arg- promastigotes vs . axenic amastigotes led to the identification of only down-regulated transcripts: an alpha tubulin ( LmxM . 13 . 0280 ) , two beta tubulins ( LmxM . 32 . 0792 , LmxM . 32 . 0794 ) , an unspecific product ( LmxM . 13 . 0300 ) and a non-coding RNA ( LmxM . 23 . ncRNA rfamscan: 218682-218827-1 ) . Furthermore , we performed a KEGG enrichment analysis , which showed a list of the top 20 regulated pathways among all samples . The list includes pathways that can be regulated in the absence of arginase activity , such as the biosynthesis of amino acids , arginine and proline metabolism and arginine biosynthesis ( Fig 4 and Table 2 ) . In this work , we focused on the L-arginine pathway and crossed the obtained data with previous metabolome fingerprints , determined by capillary electrophoresis , also focusing on L-arginine metabolism and the modulation of polyamine metabolism comparing La-WT and La-arg- promastigotes . Castilho-Martins et al . ( 2015 ) observed that the absence of arginase activity led to an increase of L-arginine and citrulline levels , but a decrease of ornithine , proline and putrescine levels . These results confirmed the importance of L-arginine supplying the polyamine pathway in L . amazonensis and also showed a possible alternative pathway to provide substrates for the pathway in the absence of arginase in the parasite [9] . In fact , to understand why the absence of arginase induced an increase in L-arginine and citrulline levels and a decrease in ornithine , proline and putrescine levels , we analyzed the transcripts levels of specific enzymes involved in these pathways , such as arginase ( LmxM . 34 . 1480/EC3 . 5 . 3 . 1 ) , pyrroline 5-carboxylate reductase ( LmxM . 13 . 1680/EC1 . 5 . 1 . 2 ) , pyrroline 5-carboxylate dehydrogenase ( LmxM . 03 . 0200/EC1 . 2 . 1 . 88 ) , glutamate 5-kinase ( LmxM . 26 . 2710/EC2 . 7 . 2 . 11 ) , spermidine synthase ( LmxM . 04 . 0580/EC 2 . 5 . 1 . 16 ) , acetylornithine deacetylase ( LmxM . 07 . 0270/EC3 . 5 . 1 . 16 ) and arginosuccinate synthase ( LmxM . 23 . 0260/EC6 . 3 . 4 . 5 ) ( Table 3 ) . Arginase transcripts were not detected in the two knockout lines ( La-arg- promastigotes and La-arg- axenic amastigotes ) , as expected . Interestingly , a reduced level of arginase was observed in La-WT axenic amastigotes , compared to promastigotes . The increase of pyrroline 5-carboxylate reductase transcripts in La-arg- compared to La-WT promastigotes showed 1 . 42-fold change . Additionally , an increase of glutamate 5-kinase transcripts was observed in both La-WT and La-arg- axenic amastigotes , compared to promastigotes . On the other hand , we observed decreased transcripts levels of pyrroline 5-carboxylate dehydrogenase , spermidine synthase , acetylornithine deacetylase and arginosuccinate synthase , with 0 . 76 , 0 . 31 , 0 . 79 and 0 . 72-fold change , respectively ( Table 3 ) . The crossing of these findings with metabolome data could explain that the increase of L-arginine and citrulline levels could be a consequence of the absence of arginase activity . The increase of L-glutamate levels could be related to the increase of pyrroline 5-carboxylase reductase transcripts . Further , as a consequence of the high consumption of L-glutamate , we observed the decrease of proline that could be related to the decrease of pyrroline 5-carboxylase dehydrogenase transcripts . The decrease of putrescine levels could be related to the decrease of spermidine synthase transcripts . The decrease of ornithine levels could be related to the decrease of acetylornithine deacetylase transcripts . And finally , the decrease of aspartate could be related to the decrease of arginosuccinate synthase transcripts ( Fig 5 ) . Additionally , we performed RT-qPCR validation of some enzymes as shown in S5 Fig . Here , we described the DE gene profile comparing the expression in promastigotes and axenic amastigotes in the presence or absence of arginase activity . In addition , we performed a correlation analysis with the KEEG arginine pathway , highlighting the regulated enzymes , as previously described in the metabolome work .
L-arginine is an amino acid used as precursor not only for protein synthesis , but also for the synthesis of NO , urea , ornithine , citrulline , creatinine , agmatine , L-glutamate , proline and polyamines [30] . On the other hand , arginase is an enzyme with regulatory roles , modulating L-arginine availability and production of ornithine , a precursor of polyamines , essential for cell replication [30] . Therefore , arginine biosynthesis is an important pathway that not only participates in the regulation of NOS2 parasite killing and arginase-mediated parasite growth , but is also involved in the regulation of the immune system [31 , 32] . The infection of murine macrophages with L . amazonensis showed increased levels of arginase I , La-arginase , arginine transporters ( CAT2B and LaAAP3 ) and miRNA modulation [32] . However , infection with La-arg- induces NOS2 expression and the production of NO , causing a lower infection index [8] and blocking miRNA expression [32] . These changes in gene regulation can indicate mechanisms to subvert the defense mechanism developed by the parasite [32 , 33] . RNA-seq technology has been used to describe transcriptomic profiles of L . major , L . mexicana and L . braziliensis [10–13 , 34] . All of these studies have provided additional knowledge about Leishmania biology and the coordinated response of Leishmania-infected macrophages in relation to gene regulation at the transcriptional level [12 , 13 , 29] . RNA-seq data are also helping to revise the previous genome annotation of L . mexicana [34] and to reconstruct some genomic regions of the L . major genome that were misassembled [35] in an attempt to improve the current genome and gene annotations . In this work , using the RNA-seq approach , we described the transcriptional profiling of L . amazonensis , compared to the phylogenetically close L . mexicana genome , since L . amazonensis genome is not completely annotated [27 , 36 , 37] . We obtained RNA-seq data from La-WT and La-arg- promastigotes and axenic amastigotes allowing the comparison of transcript abundance from different life cycle stages in the presence or absence of arginase , an important enzyme of the parasite´s polyamines synthesis . From the 8253 transcripts identified in La-WT and La-arg- promastigotes and axenic amastigotes , 60% of them were identified as hypothetical proteins and 443 were identified as novel transcripts , that did not correspond to any previously annotated genes . Although we obtained less transcripts than previously predicted for L . mexicana [34] , we could assure that the transcripts identified fulfill confidence coverage of the RNA-seq data described in this work . Recent studies have been showing the importance of characterizing a hypothetical protein not only by functional genomics , but also according to its general biological features , allowing the acquisition of new knowledge about signaling pathways , metabolism , stress response , drug resistance and in the identification of new therapeutic targets [38] . The identification of novel transcripts has been improving the accuracy of the L . amazonensis genome [34 , 35] . The finding of novel transcripts can suggest that the gene content of this organism may be higher than previously determined because the majority of novel transcripts contain open reading frames shorter than coding sequencing of genes in the current genome annotation . Another explanation could point to a novel genome organization and processing [29 , 34 , 39] . L . major chromosomes are organized as large clusters of genes or open reading frames in the same 5´- 3´ direction on the same DNA strand [40] . Each cluster of genes is processed by a transcription initiation site in a polycistronic transcription [41] . Individual mRNA transcripts are then processed , co-transcriptionally , by 5´RNA splicing and 3´polyadenylation . Therefore , without a gene specific promoter , the entire chromosome is constitutively transcribed [41] . The analysis of the transcriptome revealed 85% of genes were constitutively expressed in promastigotes and axenic amastigotes of L . amazonensis . The Leishmania genome has been described as constitutively expressed , indicating that the parasite is adapted for survival and replication in the sand-fly vector or macrophage host , using an appropriate set of genes/proteins for different environments [40] . Altogether , these results support the previous hypothesis that the Leishmania genome is mostly constitutively expressed . Interestingly , among the 1268 DE genes identified in this work , most was detected in La-arg- promastigotes ( 908 genes ) . La-arg- do not use L-arginine to produce ornithine because arginase activity is absent in this parasite line that requires polyamine supplementation for survival and replication [8] . La-arg- also presented a higher concentration of L-arginine in the cytoplasm in relation to the La-WT promastigote , probably because the parasite can sense the L-arginine pool , that leads to the regulation of L-arginine transporter expression and L-arginine uptake [42] . The absence of arginase can induce the parasite to regulate many genes involved in the arginine pathway . Therefore , the null mutant La-arg- was previously characterized and the essentiality of arginase in Leishmania in vitro growth was demonstrated with the requirement of putrescine supplementation [8] . In contrast , the arginase add-back mutant line ( La-arg-/+ARG ) restored arginase expression , growth and infectivity in vivo [8] and in vitro [32] assays . So , we focused on arginine pathway comparing our RNA-seq data with metabolome fingerprints , previously described by our group [9] . The first transcript analyzed was arginase ( EC3 . 5 . 3 . 1 ) and , as expected , no transcript was observed in both life cycle stages of the arginase knockout line ( La-arg- promastigotes and La-arg- axenic amastigotes ) , reinforcing the efficiency of the knockout methodology used [8] and indicating that the profile is maintained after amastigote differentiation . The metabolomic analysis of La-arg- promastigotes showed that absence of arginase causes an increase in L-arginine and citrulline levels , and the decrease in ornithine , putrescine and proline levels , indicating an alternative pathway to surpass the lack of this enzyme [9 , 42] . Citrulline could be metabolized by arginine deiminase ( EC3 . 5 . 3 . 6 ) or oxidoreductases ( EC1 . 14 . 13 . 39/EC1 . 14 . 13 . 165 ) . Arginine deiminase acts on carbon-nitrogen bonds [43] . Oxidoreductases act on paired donors , with incorporation or reduction of molecular oxygen on arginine biosynthesis [44] . It is interesting to note that similar transcript levels of this oxidoreductase were observed in La-WT and La-arg- promastigote and axenic amastigotes . Furthermore , the decrease in the levels of aspartate , ornithine , proline and putrescine indicates that this pathway can be used as an alternative pathway due to the differential expression of argininosuccinate synthase ( EC6 . 3 . 4 . 5 ) , acetylornithine deacetylase ( EC3 . 5 . 1 . 16 ) , pyrroline-5-carboxylate reductase ( EC1 . 5 . 1 . 2 ) , pyrroline-5-carboxylase dehydrogenase ( EC1 . 2 . 1 . 88 ) and glutamate 5-kinase ( EC2 . 7 . 2 . 11 ) . The decrease in aspartate may be due to its conversion to L-arginine succinate by argininosuccinate synthase . The decrease in ornithine could be due to the absence of L-arginine conversion in ornithine by arginase and/or the acetylornithine deacetylase consumption . Glutamate 5-kinase , which is also involved in glutamate metabolism , was not differentially expressed between La-WT and La-arg- promastigotes . This could be explained by the maintenance from the substrates L-glutamate to glutamyl-P [9] . Interestingly , increased transcript levels of glutamate 5-kinase in both La-WT and La-arg- axenic amastigotes were observed . The glutamate metabolism was previously described to be involved in the differentiation of Trypanosoma cruzi from epimastigotes to metaclyclic trypomastigotes [45 , 46] . The L-glutamate levels were increased in L-arginine deprivation , indicating a role of L-glutamate in L-arginine metabolism to supply the absence of L-arginine uptake [9] . In addition , it was described that , as alanine , L-glutamate has a key role in the cell physiology of Leishmania [47] . Other enzymes involved in the arginine pathway were not described in this work since they did not appear to be regulated in the previous metabolome analysis . In the list of top 5 transcripts differentially expressed based on the volume plot , none of the most regulated transcripts was related to the arginine pathway . However , it presented interesting DE genes . Three conserved hypothetical proteins ( LmxM . 15 . 1520 , LmxM . 08 . 0540 and LmxM . 17 . 0890 ) were identified and their characterization is important not only for functional genomics but also to improve the knowledge about signaling pathways , metabolism , the stress response , drug resistance , as well as for the identification of new therapeutic targets [38] . In addition , the identification of histone H3 and H4 ( LmxM . 06 . 0010 , LmxM . 15 . 0010 and LmxM . 10 . 0970 ) in the comparisons with La-WT ( pro La-WT vs pro La-arg- , and pro La-WT vs ama La-WT ) is indicative that histones play a central role in transcription regulation , DNA repair , DNA replication and chromosomal stability [40 , 48 , 49] in Leishmania life cycle , even in the absence of arginase activity . RNA-binding protein ( RBP ) appeared down-regulated in the comparison pro La-WT vs pro La-arg- . RBPs have been described as regulatory elements controlling the expression of genes , involving changes in mRNA stability and/or translational control . The shift from transcriptional to post-transcriptional control in trypanosomatids appears to be due to a different arrangement of protein coding genes . In addition to this specialized arrangement , the genes lack canonical promoters [40 , 50 , 51] . Currently , RBPs have been reported in Leishmania [52–54] and can elucidate gene expression regulation . Dipeptidyl-peptidase III ( DCP ) ( LmxM . 05 . 0960 ) appeared up-regulated in the comparison of La-WT vs . La-arg-axenic amastigotes . DCP belongs to the mono-zinc peptidase family . Peptidases of parasitic protozoa have been suggested as novel virulence factors , potential drug targets and vaccine candidates [55] . Previously , by microarray analyses , DCP was shown to be up-regulated in L . donovani amastigotes compared to promastigotes due to its correlation with an increase in total protease activity [56] . Thus , DCP may have a role in parasite differentiation related to nutrition and pathogenesis [56] . Similar to DCP , the protein disulfide isomerase ( PDI ) ( LmxM . 06 . 1050 ) appeared up-regulated in La-arg- axenic amastigotes compared to La-WT . PDI , a redox chaperone , has been primarily characterized with virulent and immunogenic potential [57 , 58] . Achour et al . ( 2002 ) suggested that Leishmania promastigote growth might be due to optimal protein folding as a result of the increased secretion of PDI at the surface of the parasite [58] . Later , Amit et al . ( 2014 ) showed that alanine , an inhibitor of PDI activity , caused damage to the parasite mainly in axenic amastigote forms [57] . Tryparedoxin 1 ( TXN1 ) ( LmxM . 08_29 . 1160 ) was down regulated in La-arg- axenic amastigotes compared to La-WT . TXN1 is part of the trypanothione and trypanothione reductase pathway to regulate oxidative stress [59] . The polyamine pathway can be considered metabolically important for survival and infectivity in trypanosomatids [4 , 6 , 60] . On the other hand , tryparedoxin peroxidase ( TXNPx ) ( LmxM . 15 . 1160 ) was up-regulated in La-WT axenic amastigotes compared to La-WT promastigotes , corroborating the results of previous studies demonstrating that this increase is necessary for detoxification of peroxides and resistance to NO in L . donovani , which did not show the same profile in the absence of arginase [61–63] . RNA helicases are central players in RNA biology and function . Similar to other eukaryotes , many biological functions have been attributed to trypanosomatid RNA helicases , including RNA degradation , translation regulation and RNA editing [64–66] . We identified that an ATP-dependent RNA helicase ( LmxM . 33 . 2050 ) was up-regulated in La-WT promastigotes compared to La-WT axenic amastigotes indicating a rgene regulation in Leishmania differentiation . Interestingly , the comparison of the expression profile of La-arg-promastigotes vs La-arg- axenic amastigotes showed only down-regulated genes: an α-tubulin ( LmxM . 13 . 0280 ) , two β-tubulins ( LmxM . 32 . 0792 and LmxM . 32 . 0794 ) , an unspecified product ( LmxM . 13 . 0300 ) and a ncRNA ( LmxM . 30 . ncRNA rfamscan: 218682-218827-1 ) . α-tubulin is a highly conserved protein that interacts with β-tubulin , forming an α/β-tubulin heterodimer , a key to the formation of the eukaryotic cytoskeleton , which is responsible for cell shape and it is involved in many essential processes , including cell division and ciliary and flagellar motility [67 , 68] . Altogether , these findings of down-regulated genes could be indicative of cytoskeleton reorganization dependent of the absence of arginase activity .
The transcriptional profiling of L . amazonensis reinforces the capacity of the parasite to fine-tune gene expression regulation to adapt to changes in the environment during promastigote and amastigote differentiation . It is interesting to note that although gene expression regulation in Leishmania is considered to occur at post-transcriptional levels , we observed a correlation between the transcriptomic and metabolomics data , focused on the L-arginine pathway . Additionally , the use of the arginase knockout parasite reinforced the importance of this enzyme and provided additional insights into the coordination of gene expression and parasite development and infectivity . | Leishmania are auxotrophic for many essential nutrients , including amino acids . In this way , the parasite needs to uptake the amino acids from the environment . The uptake of amino acids is mediated by amino acid transporters that are unique for Leishmania . As part of polyamine pathway , the arginase converts L-arginine to ornithine and furthermore to putrescine , products which are essential for parasite growth . On the other hand , the absence of arginase activity could alter the metabolism of the parasite to surpass the external signals during the life cycle and the fate of infection . The transcriptional profiling of La-WT and La-arg- promastigotes and axenic amastigotes revealed 8253 transcripts , 60% encoding hypothetical proteins and 443 novel transcripts . In addition , our data revealed that 85% of the genes were constitutively expressed . Among the 15% ( 1268 genes ) of the differentially expressed genes , we identified genes up- and down-regulated comparing the transcript abundance from different life cycle stages of the parasite and in the presence or absence of arginase . We also combined the transcriptional with metabolic profile that revealed a proportional correlation between enzyme and metabolites in the polyamine pathway . The differentiation of promastigotes to amastigotes alters the expression of enzymes from polyamines biosynthesis , which modulates ornithine , L-glutamate , proline and putrescine levels . In addition , the absence of arginase activity increased the levels of L-arginine , citrulline and L-glutamate and decreased the levels of aspartate , proline , ornithine and putrescine in promastigotes by differential modulation of genes involved in its metabolism . Altogether these data provided additional insights into how Leishmania is able to modulate its biological functions in the presence or absence of arginase activity to survive during environmental changes . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Conclusions"
] | [
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"gene",
"regulation",
"microbiology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"protozoan",
"life",
"cycles",
"parasitology",
"organic",
"compounds",
"developmental",
"biology",
"protozoans",
"le... | 2017 | RNA-seq transcriptional profiling of Leishmania amazonensis reveals an arginase-dependent gene expression regulation |
Cryptococcus gattii causes life-threatening disease in otherwise healthy hosts and to a lesser extent in immunocompromised hosts . The highest incidence for this disease is on Vancouver Island , Canada , where an outbreak is expanding into neighboring regions including mainland British Columbia and the United States . This outbreak is caused predominantly by C . gattii molecular type VGII , specifically VGIIa/major . In addition , a novel genotype , VGIIc , has emerged in Oregon and is now a major source of illness in the region . Through molecular epidemiology and population analysis of MLST and VNTR markers , we show that the VGIIc group is clonal and hypothesize it arose recently . The VGIIa/IIc outbreak lineages are sexually fertile and studies support ongoing recombination in the global VGII population . This illustrates two hallmarks of emerging outbreaks: high clonality and the emergence of novel genotypes via recombination . In macrophage and murine infections , the novel VGIIc genotype and VGIIa/major isolates from the United States are highly virulent compared to similar non-outbreak VGIIa/major-related isolates . Combined MLST-VNTR analysis distinguishes clonal expansion of the VGIIa/major outbreak genotype from related but distinguishable less-virulent genotypes isolated from other geographic regions . Our evidence documents emerging hypervirulent genotypes in the United States that may expand further and provides insight into the possible molecular and geographic origins of the outbreak .
Newly emerging and reemerging diseases have become a major focus of infectious disease research in the 21st century . Reemerging diseases are classified as those that have been previously documented , but are now rapidly increasing in incidence , geographic range , or both [1] . Emerging disease events have been occurring at higher than average rates in the United States due to several factors such as wildlife diversity , environmental change , international travel , and increases in host susceptibility [2] , [3] . An additional factor contributing to increases in morbidity and mortality for many infectious diseases involves genetic recombination events or gene/pathogenicity island acquisitions . These events can occur via either horizontal gene transfer or conjugation/introgression , leading to novel pathogenic genotypes . This form of virulence evolution has been well characterized in bacterial , viral , fungal , and parasitic human diseases [4] , [5] , [6] , [7] , [8] , [9] . The ability to cause damage to mammalian hosts is a common theme among all microbial pathogens , making it a key aspect of host-pathogen studies [10] . In the genomic era , it is now possible to combine conventional epidemiological approaches with newly developed molecular typing techniques to gain insight into the emergence and molecular epidemiology of pathogens . These approaches can improve understanding of population dynamics during an outbreak , and may lead to novel methods for the rapid identification , treatment , and diagnosis of emerging infections [11] . In addition , molecular typing serves as an initial approach to classify isolates into distinct genotypes for analysis . Further investigations may include the examination of virulence and phenotypic traits that may be common or distinct between genotypes [6] , [12] , [13] . Gaining insights into the molecular epidemiology and virulence of newly emerging diseases has considerable potential for the rapid assessment and management of newly emerging infections . Over the past decade , Cryptococcus gattii has emerged as a primary pathogen in northwestern North America , including both Canada and the United States [6] , [13] , [14] , [15] , [16] , [17] , [18] . In the past , C . gattii has often been associated with Eucalyptus trees in tropical and subtropical climates , causing disease in immunocompetent hosts at low incidences [19] , [20] , [21] . C . gattii is distinct from its sibling species Cryptococcus neoformans [22] , which more commonly infects immunosuppressed hosts and infects almost one million people annually with over 620 , 000 attributable mortalities [23] , [24] , [25] . C . gattii can be classified into four discrete molecular types ( VGI-VGIV ) , which represent cryptic species as no nuclear allelic exchange between groups has been observed [6] . This molecular classification is significant because VGII is responsible for approximately 95% of the Pacific Northwest infections in Canada and the United States [12] , [15] . The appearance of C . gattii in North America is alarming because this is the first major emergence in a temperate climate , indicating a possible expansion in the endemic ecology of this pathogen [26] , [27] . Several significant questions persist regarding the outbreak and its expansion within the United States . As the global collection of C . gattii isolates expands , the molecular epidemiology of the species has become increasingly informative , particularly through multilocus sequence typing ( MLST ) , which allows data to be readily compared between groups within the research community [6] , [15] , [28] , [29] , [30] . The increase in global and regional isolates that have been typed at the molecular level allows detailed analysis of C . gattii . The analysis of both conserved coding regions , and diverse noncoding regions provides insight into the genotypes responsible for the outbreak . A major finding in this study is a level of underlying diversity within the VGIIa/major genotype in the region of expansion and other geographic locales . Prior studies documented that the C . gattii VGIIa/major genotype isolates from Vancouver Island are highly virulent in experimental murine infection assays [6] . Here we expanded this analysis to examine clinical VGIIa genotype isolates from Vancouver Island , the United States , and Brazil , in addition to an environmental VGIIa isolate from California . Our findings are consistent with recent macrophage intracellular proliferation studies , demonstrating that United States isolates from the recent Pacific NW outbreak exhibit high virulence [31] . The enhanced virulence of isolates from the outbreak region , when compared with those from other regions , suggests that the genotypes circulating in the Pacific NW are inherently increased in their predilection to cause disease in mammalian hosts . In addition to the detailed examination of the VGIIa/major genotype clade , we report that the novel VGIIc genotype is highly virulent in a murine inhalation model . Moreover , the VGIIc genotype was found to have high intracellular proliferation rates in macrophages and a significantly increased percentage of mitochondria with tubular morphology after macrophage exposure , and thus VGIIc isolates share virulence attributes with the VGIIa/major genotype isolates from the Vancouver Island outbreak . These results extend the molecular and phenotypic understanding of the recently discovered VGIIc/novel genotype and help shed light into its possible geographic and molecular origins . These studies provide insights into both the evolutionary history and virulence characteristics of this unique and increasingly fatal fungal outbreak in the temperate climate of the North American Pacific Northwest and highlight the importance of a collaborative interdisciplinary approach to the analysis of emerging pathogens . Application of these approaches may increase awareness of disease risks in the expansion zone , lead to more rapid diagnoses and , as a result , accelerate the implementation of appropriate therapy .
Human and veterinary cases of confirmed or suspected C . gattii infections in the states of Washington and Oregon were identified by referring physicians and veterinarians , and subsequently isolates were purified and examined . Melanin production was assayed by growth and dark pigmentation on Staib's niger seed medium , and urease activity was detected by growth and alkaline pH change on Christensen's agar . These tests established that isolates were Cryptococcus ( C . neoformans or C . gattii ) . Isolates were concomitantly examined for resistance to canavanine and utilization of glycine on L-canavanine , glycine , 2-bromothymol blue ( CGB ) agar . Growth on CGB agar indicates that isolates are canavanine resistant , and able to use glycine as a sole carbon source , triggering a bromothymol blue color reaction indicative of C . gattii , whereas C . neoformans is sensitive to canavanine , and cannot use glycine as a sole carbon source , resulting in no growth or coloration in this selective indicator medium . All CGB positive isolates were then grown under rich culture conditions prior to storage at −80°C in 25% glycerol and genomic DNA extraction . For genomic DNA isolation , a modified protocol of the MasterPure Yeast DNA purification kit from Epicentre Biotechnologies was used . Briefly , 500 µl of glass beads ( 425–600 nm ) were added into the combination of cells and 300 µl cell lysis solution . The rest of the method followed the protocol provided by the manufacturer . For multilocus sequence typing analysis ( MLST ) [32] , each isolate was analyzed with a minimum of eight and in some cases sixteen loci . For each isolate , genomic regions were PCR amplified ( Table S1 ) , purified ( ExoSAP-IT ) , and sequenced . All primers used for the analysis were designed specifically to amplify open reading frame ( ORF ) gene sequence regions including those with non-coding DNA regions to maximize discriminatory power . Sequences from both forward and reverse strands were assembled , and manually edited using Sequencher version 4 . 8 ( Gene Codes Corporations ) . Based on BLAST analysis of the GenBank database ( NCBI ) , each allele was assigned a corresponding number . GenBank accession numbers with corresponding allele numbers are listed in the supplementary information ( Table S2 ) . To determine that the nine VGIIc/novel isolates are clonally related , given the level of diversity in the loci and the number of isolates that have been examined , we applied an equation to measure the probability of a genotype occurring more than once in the dataset [33] , [34] . For the variable number of tandem repeat ( VNTR ) analysis , the Tandem Repeat Finder ( TRF ) version 4 . 00 software package was employed for marker development , using the genomic sequence of C . gattii isolate R265 ( http://www . broadinstitute . org/annotation/genome/cryptococcus_neoformans_b . 2/Home . html ) [35] . The identified tandem repeat sequences and 400 bp of the flanking region were extracted from the genomic sequence and ranked according to the number of total repeats and the size of repeat units using an in-house Perl script ( available upon request ) . Markers were examined for stability and those with high variability and stability were chosen for the analysis . Sequences were assembled and edited using Sequencher version 4 . 8 ( Gene Codes Corporations ) and aligned using the Clustal W web based software package ( http://www . ebi . ac . uk/Tools/clustalw2/index . html ) . Mating analysis was conducted on V8 media ( pH 5 ) . Isolates were incubated at room temperature in the dark for 2–4 weeks in dry conditions . All strains were crossed with the VGIII mating type a isolate B4546 and the VGIII mating type α isolate NIH312 , both of which are fertile and commonly used for mating studies [36] . Fertility was assessed by microscopic examination for hyphae , fused clamp cells , basidia , and basidiospore formation . For each VNTR marker , a sequence type was defined as a sequence exhibiting a unique mutation . Each sequence type was confirmed to be unique by BLAST analysis of the NCBI GenBank database [37] . A concatenated VNTR sequence type ( CVST ) was defined as unique combinations of sequence types from the VNTR markers . A multiple alignment of the sequences was carried out using Clustal W software [38] . Analysis of the sequences was conducted using the Neighbor-Joining and Maximum Parsimony methods within the MEGA 3 . 1 software [39] . In addition , the use of the maximum likelihood method ( PhyML 3 . 0 ) with SH-like approximate likelihood-ratio test and HKY85 substitution model was applied [40] , [41] . For this purpose , sequences of the selected VNTR markers were concatenated . We additionally concatenated all of the strain-typing markers including the housekeeping genes used in MLST and VNTR loci for clustering analysis . The haplotype mapping analysis was carried out using TCS software version 1 . 21 ( http://darwin . uvigo . es/software/tcs . html ) [42] . A proliferation assay was previously developed to monitor the intracellular proliferation rate ( IPR ) of individual strains for a 64-hour period following phagocytosis [31] . For this assay , J774 macrophage cells were exposed to cryptococcal cells that were opsonized with 18B7 antibody for 2 hr as described previously [43] . Each well was washed with phosphate-buffered saline ( PBS ) in quadruplicate to remove as many extracellular yeast cells as possible and 1 ml of fresh serum-free DMEM was then added . For time point T = 0 , the 1 ml of DMEM was discarded and 200 µl of sterile dH2O was added into wells to lyse macrophage cells . After 30 minutes , the intracellular yeast were released and collected . Another 200 µl dH2O was added to each well to collect the remaining yeast cells . The intracellular yeast were then mixed with Trypan Blue at a 1∶1 ratio and the live yeast cells were counted . For the subsequent five time points ( T = 16 hrs , T = 24 hrs , T = 40 hrs , T = 48 hrs and T = 64 hrs ) , intracellular cryptococcal cells were collected and independently counted with a hemocytometer . For each strain tested , the time course was repeated at least three independent times , using different batches of macrophages . The IPR value was calculated by dividing the maximum intracellular yeast number by the initial intracellular yeast number at T = 0 . We confirmed that Trypan Blue stains 100% of the cryptococcal cells in a heat-killed culture , but only approximately 5% of cells from a standard overnight culture . Compared to a conventional colony counting method , this method was shown to be more sensitive in detecting the clustered yeast population or yeast cells undergoing budding . IPR values were used to assess how consistent the different VGII genotype subgroups were . For this statistical analysis the medians of each population were compared with the non-parametric Mann-Whitney U-test and values of p<0 . 025 , after controlling for multiplicity , and were accepted as statistically significant ( http://elegans . swmed . edu/~leon/stats/utest . cgi ) . The mitochondrial morphology assays were conducted in a similar way to those in previous studies , with modifications [31] . C . gattii cells , grown overnight at 37°C in DMEM in a 5% CO2 incubator without shaking for 24 hr , or isolated from macrophages 24 hr after infection , were harvested , washed with PBS twice and re-suspended in PBS containing the Mito-Tracker Red CMXRos ( Invitrogen ) at a final concentration of 20 nM . Cells were incubated for 15 min at 37°C . After staining , cells were washed in triplicate and re-suspended in PBS . For each condition , more than 100 yeast cells per replicate for each of the tested strains were chosen randomly and analyzed . For quantifying different mitochondrial morphologies , images were collected using a Zeiss Axiovert 135 TV microscope with a 100× oil immersion Plan-Neofluar objective . Both fluorescence images and phase contrast images were collected simultaneously . Images were captured with identical settings on a QIcam Fast 1394 camera using the QCapture Pro51 version 5 . 1 . 1 software . All Images were processed identically in ImageJ and mitochondrial morphologies were analyzed and counted blindly . Three individual experiments were performed for each condition and the data were tested for normality using the Shapiro-Wilk test . For homogeneity of variances we used the Levene statistic . For statistically significant differences among the mean data we applied a One-Way ANOVA . Multi-comparisons using Tukey Honestly Significant Differences tests were performed to identify statistically significant differences between pairs . A p-value of p<0 . 05 , after controlling for multiplicity , was considered to be statistically significant . Regression analysis was used to measure the correlation between tubular mitochondrial morphology and IPR values; an F-value of P<0 . 05 was considered to be a significant correlation . To examine the virulence potential of global VGII isolates , with a specific emphasis on the Pacific NW VGII outbreak genotypes , two independent murine virulence experiments were conducted at two facilities ( Duke University Medical Center and the Wadsworth Center ) . The murine virulence assays at Duke University Medical Center and the Wadsworth Center used a similar protocol to previous C . gattii and C . neoformans experimental infections [6] , [44] , [45] . At the Duke University Medical Center Animal Facility , virulence was assessed using female A/Jcr mice ( NCI , 18–24 g ) . Strains were cultured in YPD broth for 18–20 h at 30°C , harvested , washed three times with sterile PBS and counted using a hemocytometer to determine cell concentrations . Inocula for both murine experiments were confirmed by plating on YPD and counting colony-forming units ( c . f . u . ) . Nine to ten A/Jcr mice per strain were anesthetized with pentobarbital and infected via intranasal instillation with 5×104 c . f . u . in 50 µl of sterile 1× PBS . Animals that displayed severe morbidity , based on twice-daily examinations , were euthanized . Time to mortality was evaluated for statistical significance using Kaplan–Meier survival curves within the Prism software package ( GraphPad Software ) , and P values were obtained from a log-rank test . Survival data was plotted for graphical analysis using the Prism software package . At the Wadsworth center animal facility , all assays were conducted using male BALB/c mice ( approximately 6 weeks old , 15–20 g , Charles River Laboratories , Inc . ) . Strains were grown overnight in YPD broth at 30°C with shaking . The cells were harvested , washed in PBS , and counted using a hemocytometer . Five mice per strain were anesthetized with a mixture of xylazine–ketamine , and allowed to inhale 105 ( 30 µl ) cryptococcal cells per mouse , via intranasal instillation . Mice were given food and water ad libitum and monitored twice daily . At the first sign of poor health or discomfort , infected animals were euthanized . Brain and lung tissues from the dead animals were cultured on Niger seed agar for C . gattii recovery to confirm infections were due to this pathogen . Time to mortality was evaluated for statistical significance as described above . Two animals from each strain assayed in the study conducted at Duke University were selected for histopathology analysis either at the time of sacrifice or at the conclusion of the experiment for the more attenuated isolates . For each animal , lung samples were collected and stored in 10% neutral buffered formalin . Samples were paraffin embedded and hematoxylin and eosin ( H&E ) stained at the Duke University Research Histology Laboratory . After staining and slide preparation , each sample was examined microscopically for analysis of cryptococcal cell burden and immune responses . Images were captured using an Olympus Vanox microscope ( Duke PhotoPath , Duke University Medical Center ) . The animal studies conducted at the Wadsworth Center were in full compliance with all of the guidelines set forth by the Wadsworth Center Institutional Animal Care and Use Committee ( IACUC ) and in full compliance with the United States Animal Welfare Act ( Public Law 98–198 ) . The Wadsworth Center IACUC approved all of the vertebrate studies . The studies were conducted in facilities accredited by the Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) . The animal studies at Duke University Medical Center were in full compliance with all of the guidelines of the Duke University Medical Center Institutional Animal Care and Use Committee ( IACUC ) and in full compliance with the United States Animal Welfare Act ( Public Law 98–198 ) . The Duke University Medical Center IACUC approved all of the vertebrate studies . The studies were conducted in Division of Laboratory Animal Resources ( DLAR ) facilities that are accredited by the Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) .
To examine the C . gattii outbreak isolates collected from 2005 to 2009 ( Figure 1 ) , an in-depth stepwise molecular analysis was applied to each isolate , and the genotypes were compared with other global genotypes . In total , 20 markers were selected for analysis . These markers include both coding and noncoding genomic regions and range in size and allelic diversity ( Table 1 ) . Additionally , all of the markers are randomly distributed among the chromosomes in the most recent assembly of the reference C . gattii VGI genome , WM276 ( Figure 2 ) . Initially , all isolates were sequenced at a total of eight MLST markers , and four variable number of tandem repeats ( VNTR ) markers ( Figure 3 , Table 2 ) . Next , global isolates were selected for diversity , and several isolates from each of the primary genotypes in the expansion region were chosen for sequence analysis at eight additional MLST loci , bringing the total number of genetic markers analyzed for these isolates to 20 ( Figure 4A ) . As expected , the MLST markers were less variable and more conserved , while the VNTR markers allowed for higher-resolution differentiation between isolates that appeared identical by MLST analysis . The generated datasets were then concatenated both without and with VNTR data ( Figure 4B , Figure 4C ) . The combined analysis of the results presented here , and a 30 marker MLST analysis conducted previously [6] , [18] , reveal several findings of interest in relation to VGII genotypes in the region . From the analysis of 34 markers ( 30 MLST/4 VNTR ) , we show that the Vancouver Island VGIIa/major isolates are fully identical at all loci to several recent isolates from Washington and Oregon , as well as a historical clinical isolate ( 1970's ) , NIH444 , from Seattle . Additionally , the VGIIb/minor isolates from Australia and Vancouver Island are identical at 34 total loci , and also identical to VGIIb/minor isolates from Oregon at 20 loci ( 16 MLST/4 VNTR ) . Furthermore , all VGIIc isolates to date are identical across all 20 loci examined ( Figure 4A ) . However , we also are able to discriminate the outbreak VGIIa genotype from an environmental VGIIa isolate from California , CBS7750 , and clinical VGIIa isolates CA1014 and ICB107 from California and Brazil , respectively , at one or more MLST/VNTR loci . It is clear from prior studies that the VGIIa/major and VGIIb/minor isolates are clonal lineages [6] , [12] , [15] , [46] , and here we confirmed that this is the case for the nine VGIIc/novel isolates , based on 7-loci MLST analysis of the global VGII population ( Figure S1 ) ( p<0 . 0001 ) . The largest and most comprehensive dataset arose from the combined analysis of seven MLST and four VNTR loci , resulting in a total of 41 sequence types ( STs ) . This dataset was generated from clinical , veterinary , and environmental C . gattii isolates ( Figure 3 , Figure S1 , Table S3 ) . From the analysis , it is clear that the VGIIa/b/c clusters are all related to each other , but also distinct . In addition , the data show that the VGIIa/major clade is closely clustered to VGIIc , further validating prior reports that examined a more limited number of loci [13] , [47] . In addition , VGIIc ( ST21 ) shares high sequence identity to ST34 , represented by a mating type a clinical isolate from Colombia , suggesting that the VGIIc genotype may have resulted from a-α mating , even though all isolates related to the Pacific NW outbreak are exclusively α mating type . Additionally , Vancouver Island isolates from our collection that had not been fully typed by MLST were sequenced at two loci to determine if any were unrecognized VGIIc isolates ( n = 56 ) ( Figure S2 ) . Of these , 51 were found to be VGIIa , five were VGIIb , and none were VGIIc , consistent with previous data from the region . Thus , VGIIc appears to remain exclusive to the United States , specifically Oregon , and has never been reported from Vancouver Island , the mainland of Canada , Washington State , or elsewhere globally . Within the VGIIa/major cluster , based on the initial MLST analysis of 30 loci , only a single isolate ( ICB107 ) could be distinguished from the other VGIIa isolates , and this was at only one locus [18] . To further investigate this homogeneous population causing the vast majority of the outbreak-related morbidity and mortality , we expanded the molecular analysis to include highly variable regions of the genome . The application of these VNTR markers , in combination with the MLST markers , allowed us to generate five independent STs from within the VGIIa/major genotype and related isolates ( Figure 3 ) . These five sequence types ( ST1 , ST2 , ST3 , ST13 , ST30 ) contained a total of 44 isolates ( Figure 3 , Table S3 ) . The canonical VGIIa/major outbreak genotype , ST1 , contained the vast majority of the 44 isolates ( n = 38 ) . As expected based on previous models of the C . gattii outbreak expansion [13] , ST1 consisted of isolates exclusively from the initial outbreak and expansion zones , including British Columbia , Washington , and Oregon ( Table S3 ) . These results further validate the hypothesis that the epicenter of the outbreak was on Vancouver Island , beginning in the late 1990's , with a direct expansion into neighboring mainland British Columbia and subsequently into the United States [13] . The only exception in this dataset is isolate NIH444 , an older isolate from the region that was isolated from a patient sputum sample in Seattle in the early 1970's [18] , which is also identical at all 34 markers examined . This suggests that the VGIIa/major genotype responsible for most of the outbreak cases may have been circulating in the region prior to the outbreak . The possible travel history of this patient is unknown , and could therefore have involved exposure on Vancouver Island . Overall , this analysis provides increased evidence that the outbreak genotype is unique to the region thus far , and molecularly distinct from closely related isolates from both California and South America . While the homogeneous nature of the VGIIa/major isolates based on robust molecular typing validated previous models , an underlying diversity within this group was also discovered . First , we further validated that the isolate ICB107 ( ST13 ) , from Brazil , was indeed distinct from the ST1 VGIIa/major clade . This isolate differs at one MLST marker ( LAC1 ) , and three VNTR markers ( VNTR3 , VNTR15 , VNTR34 ) . Additionally , the high-resolution sequence analysis was able to discriminate other VGIIa isolates that were collected from California . These include isolate CBS7750 ( ST3 ) , collected from the environment in San Francisco in 1990 [48] , and isolate CA1014 ( ST2 ) , which was isolated from a patient with HIV infection in southern California . Each of these two isolates differs from ST1 due to unique mutations within the VNTR7 and VNTR34 loci , respectively . This shows that similar VGIIa genotype isolates have been found elsewhere , but that none are identical to those circulating as part of the ongoing Vancouver Island outbreak . Whether these isolates are a result of drift from ST1 , or if ST1 arose from one of these related genotypes is not known . In addition to discriminating VGIIa isolates that were not from the outbreak region , we also found a novel ST , ST30 , which is highly similar to ST1 , but divergent at a unique region of VNTR34 . Interestingly , all three of the ST30 isolates are exclusively from Oregon , including two human clinical cases and one marine mammal case ( Figure 1 , Figure 3 , Table S3 ) . These results are consistent with an expansion followed by genetic drift in the highly variable VNTR loci . Isolates of ST30 have not been detected on Vancouver Island , indicating that this divergence is recent , and likely occurred after the expansion of ST1 into the United States . Alternatively , both ST1 ( VGIIa/major ) and ST30 may have been present for a long period , with only ST1 having been transferred to Vancouver Island . To gain insights into the potential origins of the VGIIc genotype , and to assess its position within the overall VGII clade , clustering analysis was applied . Analysis of the combined dataset including 41 sequence types generated from 115 C . gattii isolates shows that the VGIIc genotype is independent , but similar to VGIIa ( Figure 3 ) . The closest relationship determined from the analysis was to ST34 , an isolate from Colombia , which is also of the opposite a mating type . Moving beyond the direct branch , it appears that the VGIIc genotype shares sequence similarities to global isolates from South America , Africa , and also European isolates with likely African origins based on collected clinical case histories . Additionally , the VGIIc group also shares the IGS1 allele with isolates from Australia , further obscuring the possible origins and necessitating a more thorough analysis ( Figure 4A ) . When the clustering analysis was expanded to include additional MLST loci ( Figure 4A ) , both with and without the VNTR markers , the relationships of VGIIc to other global genotypes was further elucidated , with close relationships observed with global isolates from South America , Africa , Europe ( Greece ) , and Australia ( Figure 4B , Figure 4C , Table S4 ) . These results increase the comprehensiveness of the analysis , and allow predictions of the relationship of this genotype to global isolates . Examination of alleles illustrates that , when the analysis is expanded , the VGIIc group appears to be more diverse from VGIIa and VGIIb . Each allele represented in green was initially denoted as an allele that was unique to the VGIIc genotype , with a total of seven such alleles ( Figure 4A ) . To further elucidate the possible origins of these alleles , isolates selected based on their global diversity were sequenced at these loci ( Figure 4A ) . Identical matches for four of the seven VGIIc-unique alleles were identified in isolates from Brazil , Australia , Europe , and European isolates with likely African origins , while three alleles ( SXI1α , HOG1 , and CRG1 ) remain unique to this novel genotype and only seen in Oregon thus far ( Figure 4A ) . To further characterize the genetic relationships among the global isolates in relation to the outbreak isolates , maximum likelihood ( ML ) analysis was applied . Initially , the isolates were characterized at 15 MLST loci , excluding the MAT locus so that both α and a isolates could be included . This analysis indicates that VGIIc may be more distantly related to the VGIIa/major genotype than initially observed . In addition , analysis of the 15 MLST loci shows a possible relation of VGIIc with isolates from South America , Africa , Europe , and Australia ( Figure 4B ) . When this analysis was expanded to also include the four VNTR loci , similar results for the global comparisons of all genotypes and the relation of VGIIc to global isolates were observed ( Figure 4C ) . For these reasons , additional sampling and analysis will be necessary to more precisely elucidate if this novel virulent genotype originated locally , or originated in an under-sampled region . In addition to clustering analyses , TCS haplotype-mapping software was applied to establish the evolutionary histories of the MLST alleles examined during the analysis ( Figure 5 , Figure 6 , Figure S3 ) . From the sequence results , all of the VGIIc isolates were determined to be 100% identical , indicating that there was likely a recent emergence in which all of the isolates are clonally derived . To test this hypothesis , the TCS analysis allowed for the examination of individual loci to determine which alleles are likely ancestral , intermediate , or recently derived . Of the sixteen loci examined , eight were consistent with VGIIc possessing the ancestral allele , six of the alleles were distal nodes at the terminal end of the respective haplotype networks , and two loci were of intermediate allele positions . Alleles with ancestral genotypes are less informative because these alleles may not have diversified over time in the VGIIc lineage for various reasons , including selection pressures and overall lack of diversity at the allele . When only non-ancestral alleles were examined , 75% lay at the distal ends of their haplotype maps . Intriguingly , the three VGIIc alleles unique to the genotype ( SXI1α , HOG1 , and CRG1 ) all have distal placements ( Figure 5A–C ) . Additionally , the most recent ancestor to VGIIc in all three cases can be shown to derive from isolates that are from South America and Australia , indicating that VGIIc may have emerged out of one of these regions ( Figure 5 ) . While other regions including Europe and North America can be seen , no other regions are observed for all three of these alleles . These distal placements are consistent with a recent divergence of the unique VGIIc lineage . The haplotype analysis , in combination with the lack of any underlying diversity within the nine VGIIc isolates analyzed , indicates a recent emergence of this novel virulent genotype in Oregon . To examine the role that recombination may have played in the population structure of the VGII molecular type , we conducted paired allele analysis for 25 representative global isolates ( Figure 6 , Figure S4 ) . The discovery of all four possible allele combinations between two unlinked loci ( AB , ab , Ab , aB ) serves as evidence for likely recombination [49] . From this analysis , we show that isolates collected from South America , Africa , and Australia appear to be involved in recombination events . Representative VGIIa/major , VGIIb/minor , and VGIIc/novel isolates were found among groups of recombinant isolates . A group of ten isolates , all α , from South America and Africa ( Figure S4 ) appeared most commonly as recombinant partners , although several a mating type isolates were also less frequently involved . In further support , when we examined the number of genotypes present by region and compared this data to the total number of genotypes represented ( Figure S1 ) , it is clear that South America and Africa populations are more diverse when compared with isolates from North America , which are more clonal . Additionally , while the observed diversity in Australia was lower than South America and Africa , this may be attributable to sampling bias of clonal regions as prior studies have shown that this continent is a region with high levels of recombination due to both same-sex and opposite-sex mating events [50] . In addition to the paired allele analysis , allele diagrams were constructed to observe possible recombination within individual MLST loci ( Figure S5 ) . The most parsimonious explanation for allelic diversity in 11 of the MLST loci analyzed is as a result of consecutive and/or independent mutations within the population . Within the four remaining loci , there exists at least one hybrid allele that may be the result of a recombination event between two hypothesized parental alleles in the global VGII population ( Table 3 , Figure S5 ) . Phenotypic mating results were conducted and illustrate that the VGIIa/major ( α ) , VGIIc/novel ( α ) , VGII mating type a genotypes , as well as several of the proposed parental contributors from the allelic and genotypic recombination analysis show fertility with the production of spores when mated with fertile VGIII isolates ( Table S5 ) . Taken together , this suggests that both α-α and a-α mating events may be contributing to the formation of recombinant genotypes as well as the production of infectious spores . There were no examples of alleles introgressed into VGII from VGI , VGIII , or VGIV , in accord with findings that the four VG molecular types likely represent cryptic species [6] , [29] . In summary , these results suggest that recombination events may be critical driving forces in the evolution of C . gattii VGII diversity , which may in part contribute to the generation of genotypes displaying increased virulence . It has recently been shown that intracellular proliferation rate ( IPR ) values for cryptococcal cells within macrophages are positively correlated with virulence in the murine model for cryptococcosis [31] . To further elucidate the potential virulence of outbreak isolates collected from the United States , proliferation rates of selected isolates were tested and compared to other isolates for which proliferation data had been previously obtained . In total , IPR values for eight of the nine VGIIc isolates were measured ( Figure 7A ) . In addition , the type strains for VGIIa/major ( R265 ) and VGIIb/minor ( R272 ) were included as controls , and previously published data for other VGIIa and VGIIb isolates were included for comparisons [31] . On the basis of individual strains , seven of the eight VGIIc/novel isolates showed high IPR levels , with only a single outlier ( EJB52 ) that had a low IPR value ( 0 . 97 ) . Taken together , the median IPR value for VGIIc is significantly closer to that of VGIIa/major than to VGIIb/minor ( Figure 7A ) . These results indicate that the VGIIc genotype has a similar intracellular phenotype , and thus virulence profile to the VGIIa/major genotype . This is noteworthy because previous analysis showed that the VGIIa/major genotype isolates from the outbreak had unusually high IPR values , and the VGIIc isolates from the same outbreak are here shown to have similarly high IPR values . Another unique feature of the outbreak VGIIa/major isolates is the ability to form highly tubular mitochondria after intracellular parasitism , a characteristic that correlates with both IPR and murine virulence [31] . To explore the morphology of VGIIc isolates , we examined selected isolates in DMEM media and after exposure to macrophages . This analysis included two VGII environmental isolates ( CBS8684 , CBS7750 ) and four of the VGIIc/novel isolates . As expected , the vast majority of the mitochondria for all six isolates were non-tubular after exposure to DMEM media alone ( Figure 7B ) . However , after exposure to macrophages , three of the four VGIIc isolates tested showed significantly higher percentages of tubular morphology ( Figure 7C ) . The lone VGIIc isolate that did not exhibit this morphology ( EJB52 ) was the same isolate that also had a low IPR value , and is thus an overall outlier for the VGIIc genotype . When the results of IPR versus percentage of cells exhibiting tubular morphology were plotted , the graph showed a statistically significant correlation of the two measures with an R2 value of 0 . 85 ( Figure 7D ) . These results further indicate that the VGIIc genotype is phenotypically similar to the Vancouver Island VGIIa/major outbreak strains . Our results also support evidence for similar mechanisms regulating the increased virulence seen in the novel VGIIc genotype . The exact roles that the mitochondrial tubular morphology might play in virulence are not yet known . However , the distinct phenotype is clearly unique to the outbreak isolates and is correlated with an increased ability to grow and divide within host innate immune cells . The VGIIc isolates were found to be highly virulent in the murine inhalation model of infection . Two studies were conducted to examine virulence . In the first murine experiment a total of six isolates ( n = 5 animals/isolate ) , were examined including two VGIIc isolates ( Figure 8A ) . The VGIIa/major isolate R265 served as a positive control for high virulence , based on prior studies [6] , and the VGIIc isolates EJB15 and EJB18 showed similar virulence with this well characterized virulent isolate . Additionally , two VGIIa isolates that are not hypothesized to be from the current Vancouver Island outbreak , including NIH444 , which is fully identical across 34 markers , and isolate CA1014 , which differs from R265 at VNTR34 , show a significant reduction in virulence compared to the high virulence isolates ( P<0 . 05 ) . Finally , in accordance with previous studies , the VGIIb/minor type strain R272 from Vancouver Island was avirulent in this model . The analysis of virulence within the VGII genotype was extended in a second experiment , in which 12 isolates ( n = 9–10 animals/isolate ) were examined . This study included two VGIIa/major isolates from the outbreak zone , two VGIIb/minor isolates from the outbreak zone , five of the novel VGIIc isolates , two VGIIa-related isolates that are not part of the outbreak , and the C . neoformans var . grubii type strain , H99 . The H99 isolate used ( H99S ) has been shown to be highly virulent in the murine model of infection [44] , [51] . As expected , all five of the VGIIc isolates from Oregon as well as the VGIIa/major isolates from Vancouver Island and Oregon , and the highly virulent H99 isolate exhibited a high level of virulence ( median survival = 20 . 6 days ) . The VGIIb/minor isolates tested were significantly decreased in virulence compared to the more virulent VGIIa and VGIIc genotypes ( P<0 . 005 ) . The VGIIb isolate R272 was avirulent whereas the VGIIb isolate EJB53 from Oregon exhibited significantly less virulence compared to the VGIIa/major and VGIIc isolates ( P<0 . 005 , median survival = 46 days ) . Similar to the first animal study , two VGIIa isolates that differ at one or more molecular markers from the major VGIIa outbreak genotypes were also tested . The environmental isolate CBS7750 and a clinical isolate from South America ICB107 were significantly attenuated ( P<0 . 005 ) ( Figure 8B ) . These results provide further evidence that these are related to but distinguishable from isolates that are specific to the Vancouver Island outbreak , and subsequent United States expansion , and are decreased in ability to mount fatal infections in a mouse intranasal instillation model of infection . The cause of infection was further evaluated by histopathological analysis of lung sections recovered from two infected animals per isolate at sacrifice . Harvested organs were processed and sectioned for slides with H&E staining . The lungs from the virulent isolates showed significant inflammation and numerous cryptococcal cells dispersed throughout the alveoli , in accordance with severe pulmonary infection . Our findings show that there are no major clinical differences between pulmonary infections with the infectious genotypes VGIIa/major ( Figure 8C ) , and the novel VGIIc genotype ( Figure 8D ) . These results further support similar disease progression caused by these two highly virulent outbreak genotypes .
The findings presented here document that the outbreak of C . gattii in Western North America is continuing to expand throughout this temperate region , and that the outbreak isolates in the United States of both the VGIIa/major genotype and the novel VGIIc genotype are clonally derived and highly virulent in host models of infection . These conclusions are based on an extensive molecular analysis of isolates collected from the United States ( Table 2 ) and a comprehensive global collection of VGII isolates of diverse geographic origin ( Figure S1 ) , examining both conserved and divergent regions of the genome . The virulence analysis is based on assays in both murine derived macrophages and mice . These findings demonstrate that this emerging and fatal outbreak is continuing to expand , and that the virulence of these isolates is unusually high when compared to isolates of closely related but distinguishable genotypes found in other non-outbreak regions . The continued expansion of C . gattii in the United States is ongoing , and the diversity of hosts increasing . Cases have been observed in urban and rural areas , and have occurred in a range of mammals [16] , [52] . On Vancouver Island and the mainland of British Columbia , cases have been documented in marine and terrestrial mammals including cats , dogs , porpoises , ferrets , and llamas [15] , [52] , [53] . This trend has continued in the United States , with several cases in agrarian , domestic , and wild terrestrial mammals , as well as marine mammals , adding elk , alpacas , and sheep to the aforementioned list ( Table S1 ) [13] , [14] , [17] . The co-expansion of the outbreak among mammals and humans is significant for several reasons . Non-migratory mammals serve as sentinels for disease expansion , particularly given that isolation of C . gattii from the environment is difficult , and not yet successful at all in Oregon . Additionally , the threat to agricultural and domestic animals is significant and thus the need for cooperation among health officials is critical . Finally , the widespread spectrum of disease illustrates that the organism is likely to be pervasive in the environment , and that physicians and veterinarians should be well informed of symptoms to facilitate early diagnoses , and successful isolate collection and tracking . A major question in the study of this outbreak is whether sexual recombination , either within or between mating types , is occurring or has occurred in the region . The possibility of meiosis is important for two reasons . The first is that sexual recombination is postulated to be a driving force for the increased virulence of the VGIIa/major genotype , supported by the discovery of a diploid VGIIa/major isolate , an intermediate in unisexual mating ( all nine VGIIc/novel isolates are haploid ) [6] , [36] . C . gattii has also been shown to undergo opposite sex mating in the laboratory , although this has not yet been observed to occur between two isolates of the VGII molecular type [36] , [54] , [55] . Studies in C . neoformans have shown that this related pathogen completes a full a-α sexual cycle in association with plants [56] . Additionally , a recent study of environmentally sampled Australian VGI isolates demonstrated evidence for recombination via both opposite and same-sex mating [50] . Taken together , available evidence indicates that both opposite and same-sex mating are naturally occurring in populations . This evidence lends support to the hypothesis that meiosis might be a factor in the forces that are driving high virulence in the outbreak region . The second major event that results from sexual processes in the pathogenic Cryptococcus species is the formation of spores . Small spores ranging from 1–2 µm in diameter have been observed to be produced in large numbers as the result of opposite sex mating in both C . neoformans and C . gattii [57] , [58] . Studies by Lin and colleagues showed that sexual spores can be produced as the result of a meiotic process occurring between cells of the same mating type , a process referred to as unisexual or same-sex mating [59] . Several studies have shown spores to be pathogenic in animal models of infection . Two previous studies both showed evidence for virulence of Cryptococcus spores , and in one case provided evidence for enhanced virulence compared to yeast cells [60] , [61] . More recently , studies have shown that Cryptococcus neoformans spores are indeed virulent in the murine intranasal instillation model of infection [44] , [62] , providing evidence that spores should be considered as infectious propagules in models examining infections , expansion , and emergence of both C . neoformans and C . gattii . Given that all of the Pacific NW isolates are α mating type , and particles small enough to be spores are present in the air [26] , [63] , the most parsimonious model is that if these are spores , they are produced via α-α unisexual reproduction . Our findings further indicate that mitochondria may play a significant role in the increased virulence seen in the outbreak isolates [31] . Tubular morphology and the increased ability to proliferate within immune cells indicate that the ability to proliferate and survive within host cells is fundamental to virulence . The possible role of mitochondrial involvement is intriguing and also increasingly relevant based on studies that have shown mitochondrial inheritance and recombination may impact C . gattii evolution , with the inheritance of the mitochondrial genome from the a mating type parent in opposite-sex mating [64] , [65] . Future studies in this area should address the roles that mitochondrial genes , or nuclear genes that regulate mitochondria may play in the hypervirulence observed in the outbreak isolates . Furthermore , it may be that cell-cell fusion events via mating and mitochondrial exchange without meiosis or nuclear genetic exchange have played roles in recombination and virulence acquisition in naturally occurring C . gattii populations [64] , [65] . A central question in the field lies in the possible origins of the virulent genotypes . For the VGIIa and VGIIc lineages , it is clear that those are unique to the Pacific NW , and either arose there locally , or were transferred from an under-sampled region ( Australia , South America , Africa ) . Isolates that are related to , but distinct at one or more molecular marker from VGIIa have been identified in San Francisco ( CBS7750 ) , southern California ( CA1014 ) , and South America ( ICB107 ) . However , in each of these cases , the isolates are not identical with the VGIIa/major isolates from the Pacific NW . Whether the outbreak isolates are derived from these isolates , or alternatively that these isolates are derived from the outbreak lineage is at present unclear . In the VGIIb/minor outbreak lineage , isolates from Australia are identical at all 30 MLST loci and four VNTRs analyzed , and the most parsimonious model is that the two are directly related . While it is conceivable that both the Australian and the Vancouver Island VGIIb/minor genotype isolates were dispersed independently from another geographic locale , until isolates are identified conclusively from another locale the most parsimonious model is transfer from Australia to the Pacific NW . We note that a single isolate with a related but distinct genotype ( isolate 99/473 ) from the Caribbean has been identified; and other isolates have been reported to share the VGIIb genotype but have been analyzed at a limited number of MLST markers ( n = 7 ) which is insufficient to establish how closely related these isolates are to the outbreak VGIIb/minor genotype strains [29] . The origins of VGIIc are unclear , with the genotype possibly arriving in the Pacific NW from South America , Africa , Europe , or Australia . Alternatively , this novel unique genotype may have arisen locally . As for the geographic origins of VGII diversity , this also remains to be established and may involve populations in Australia , South America , and Africa . It is clear that there is considerable diversity among isolates from South America . As we originally proposed as an alternative model [6] , and has been independently presented by other investigators ( W . Meyer , T . Boekhout , JP Xu , pers . comm . ) , South America may represent a source of diversity and ongoing generation of novel isolates . Analysis of 8 MLST loci in this study indicates that in South America and the Caribbean there are 14 genotypes seen in 21 isolates , while in North America only 3 genotypes have been observed through the analysis of 64 isolates ( Figure S1 ) . Additionally , there is accumulating evidence that fertile isolates of both a and α mating type are present in South America [29] , and thus ongoing a-α opposite sex mating may be occurring there . It is also clear that a unique set of VGII isolates are circulating in Australia , and there is evidence for ongoing recombination in α only and a-α populations , suggesting that mating contributes to the generation of diversity in Australia [36] , [49] , [54] , [55] , [66] , [67] . Finally , the analysis of global VGII isolates reveals genetic diversity in Africa , and given the recent findings that C . neoformans likely originated in sub-Saharan Africa ( A . Litvintseva and T . Mitchell , pers . Comm . ) , further analysis of African C . gattii isolates is clearly warranted . It remains possible that South America , Africa , or both represent the ancestral populations of C . gattii , and that more recent dispersal events from other established populations ( for example , from Australia to the Pacific Northwest ) have occurred to contribute to the outbreak . As yet , all of the isolates found in the Pacific Northwest are α mating type . Thus , if sexual reproduction is occurring in the Pacific Northwest , it would appear to involve same-sex mating occurring under environmental conditions . Recent studies have documented that C . neoformans and C . gattii are stimulated to undergo opposite-sex mating in laboratory conditions that simulate environmental niches ( pigeon guano medium , co-culture with plants ) and thus similar conditions may be necessary in nature [56] , [68] . Overall , both the VGIIa/major and the VGIIc/novel genotypes contain a number of MLST loci that are thus far restricted to these lineages , and their origins remain to be identified . Independently of the variables leading up to and influencing this outbreak , the major concern is and continues to be the inexorable expansion throughout the region . From 1999 through 2003 , the cases were largely restricted to Vancouver Island . Between 2003 and 2006 , the outbreak expanded into neighboring mainland British Columbia and then into Washington and Oregon from 2005 to 2009 . Based on this historical trajectory of expansion , the outbreak may continue to expand into the neighboring region of Northern California , and possibly further . The rising incidence of cryptococcosis cases in humans and animals highlights the need for enhanced awareness in the region , and those regions that may potentially become involved . While rare , little is currently known about how or why specific humans and animals become infected . Increased vigilance may decrease the time from infection to diagnosis , and thus lead to more effective treatment and a reduction in mortality rates . The potential dangers of travel-associated risks should be noted , as a growing number of cases attributable to travel within the Pacific NW region have been documented [69] , [70] . Northern California has similar temperate climates to endemic regions within Oregon , leading to the hypothesis that the emergence may expand there , while expansion eastward may be limited by winters with average temperatures often below freezing [17] . The expansion of the outbreak into California is plausible based on several studies documenting the presence of C . gattii throughout the state and in Mexico . C . gattii molecular type VGII was environmentally isolated in the San Francisco area in 1990 ( isolate CBS7750 ) [48] , and there have also been two confirmed and one travel-associated case of C . gattii molecular type VGI in California . Of the VGI cases , one occurred in a male Atlantic bottlenose dolphin in San Diego , one was isolated from a liver transplant recipient in San Francisco , and the other from an otherwise healthy patient in North Carolina with travel history to the San Francisco region [71] , [72] , [73] . In addition C . gattii has been reported in southern California among a cohort of HIV/AIDS patients [74] . Recently , studies of clinical isolates from Mexico revealed all four molecular types of C . gattii to be present [75] . Taken together , the hypothesis that the virulent isolates from the Pacific NW will expand into California must be considered by both physicians and public health officials . During the coming years , monitoring and researching the outbreak expansion as a multidisciplinary effort will be critical . The ability to bring diverse groups of professionals interested in C . gattii expansion has been greatly facilitated through the formation of the Cryptococcus gattii working group of the Pacific Northwest [17] . From a research standpoint , further examination of the molecular mechanisms underlying the increased virulence in both VGIIa/major and VGIIc/novel will be useful for the development of aggressive treatments that may be needed . Furthermore , increased efforts to determine the ecology and population dynamics of C . gattii in the region , and elucidating the evolutionary history of the VGIIc genotype will be critical to gain further insights into the origins of this unprecedented and frequently fatal fungal outbreak . | Emerging and reemerging infectious diseases are increasing worldwide and represent a major public health concern . One class of emerging human and animal diseases is caused by fungi . In this study , we examine the expansion on an outbreak of a fungus , Cryptococcus gattii , in the Pacific Northwest of the United States . This fungus has been considered a tropical fungus , but emerged to cause an outbreak in the temperate climes of Vancouver Island in 1999 that is now causing disease in humans and animals in the United States . In this study we applied a method of sequence bar-coding to determine how the isolates causing disease are related to those on Vancouver Island and elsewhere globally . We also expand on the discovery of a new pathogenic strain recently identified only in Oregon and show that it is highly virulent in immune cell and whole animal virulence experiments . These studies extend our understanding of how diseases emerge in new climates and how they adapt to these regions to cause disease . Our findings suggest further expansion into neighboring regions is likely to occur and aim to increase disease awareness in the region . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/fungal",
"infections",
"molecular",
"biology",
"genetics",
"and",
"genomics",
"microbiology"
] | 2010 | Emergence and Pathogenicity of Highly Virulent Cryptococcus gattii Genotypes in the Northwest United States |
Humans and animals are able to learn complex behaviors based on a massive stream of sensory information from different modalities . Early animal studies have identified learning mechanisms that are based on reward and punishment such that animals tend to avoid actions that lead to punishment whereas rewarded actions are reinforced . However , most algorithms for reward-based learning are only applicable if the dimensionality of the state-space is sufficiently small or its structure is sufficiently simple . Therefore , the question arises how the problem of learning on high-dimensional data is solved in the brain . In this article , we propose a biologically plausible generic two-stage learning system that can directly be applied to raw high-dimensional input streams . The system is composed of a hierarchical slow feature analysis ( SFA ) network for preprocessing and a simple neural network on top that is trained based on rewards . We demonstrate by computer simulations that this generic architecture is able to learn quite demanding reinforcement learning tasks on high-dimensional visual input streams in a time that is comparable to the time needed when an explicit highly informative low-dimensional state-space representation is given instead of the high-dimensional visual input . The learning speed of the proposed architecture in a task similar to the Morris water maze task is comparable to that found in experimental studies with rats . This study thus supports the hypothesis that slowness learning is one important unsupervised learning principle utilized in the brain to form efficient state representations for behavioral learning .
The nervous system of vertebrates continuously generates decisions based on a massive stream of complex multimodal sensory input . The strength of this system is based on its ability to adapt and learn suitable decisions in novel situations . Early animal studies have identified learning mechanisms that are based on reward and punishment such that animals tend to avoid actions that lead to punishment whereas rewarded actions are reinforced . The study of such reward-based learning goes back to Thorndikes law of effect [1] . Later , the mathematically well-founded theory of reinforcement learning , which describes learning by reward , has been developed [2] , [3] . In a general reinforcement learning problem , an agent senses the environment at time via a state , where is the state space of the problem . The agent then chooses an action , which leads to state according to some ( in general probabilistic ) state-transition relation . The agent also receives some reward signal , which depends probabilistically on the state . By choosing an action the agent aims at maximizing the expected discounted future rewardwhere denotes the expectation and is some discount rate . This general theory has a huge influence on psychology , systems neuroscience , machine learning , and engineering and numerous algorithms have been developed for the reinforcement learning problem . By utilizing these algorithms , many impressive control applications have been developed . Several experimental studies connect the neural basis for reward-based learning in animals to well-known reinforcement learning algorithms . It has been shown that the activity of dopaminergic neurons in the ventral tegmental area is related to the reward-prediction error [4] , a signal that is needed for parameter updates in temporal difference learning [3] . These neurons in turn have dense diffuse projections to several important areas including the striatum . In the striatum it was shown that dopamine influences synaptic plasticity [5] . Hence , the principal basis of reward-based learning in this sub-system , although not well understood yet , could be related to well-known reinforcement learning algorithms . However , the learning capabilities of animals such as rodents are still far from reach with current reinforcement learning algorithms . Since physiological experiments are consistent with quite standard reward-based learning schemes , it is reasonable to speculate that the superior learning capabilities of animals is to a high degree based on the ability to autonomously extract relevant features from the input stream such that subsequent reward-based learning is highly simplified ( We note that the distinction between feature extraction and reward-based learning is most likely not so strict in the brain . For example , acetylcholine is a prominent neuromodulator in sensory cortical areas which could be utilized for task-dependent feature extraction ) . In fact , one of the most crucial design questions in the design of a reinforcement learning system is the definition of the state space . Most reinforcement learning algorithms are only applicable if the state space of the problem is sufficiently small . Thus , if the sensory input to a controller is complex and high-dimensional , the first task of the designer is to extract from this high-dimensional input stream a highly compressed representation that encodes the current state of the environment in a suitable way such that the agent can learn to solve the task . In contrast , the nervous system is able to learn good decisions from high-dimensional visual , auditory , tactile , olfactory , and other sensory inputs autonomously . The autonomous extraction of relevant features in the nervous system is commonly attributed to neocortex . The way how neocortex extracts features from the sensory input is still unknown and a matter of debate . Several principles with biologically plausible neural implementations have been postulated . Possible candidates are for example principal component analysis ( PCA ) [6] , [7] , independent component analysis [8]–[10] , and information bottleneck optimization [10] , [11] . One learning algorithm that exploits slowness information is slow feature analysis ( SFA ) [12] . SFA extracts the most slowly varying features in the input stream ( see below ) . One important property of SFA is that it can be applied in a hierarchical fashion , first extracting local features on the raw input data which are then integrated to more and more global and abstract features . This hierarchical organization is similar to cortical organization for example in the visual system ( we note however that the characteristic recurrent organization of cortex where multiple loops provide feedback from higher-level to lower-level processing is not yet exploited in hierarchical SFA architectures ) . Furthermore , the features that emerge from SFA have been shown to resemble the stimulus tunings of neurons both at low and high levels of sensory representation such as various types of complex cells in the visual system [13] as well as hippocampal place cells , head-direction cells , and spatial-view cells [14] . These features have been extracted from visual input . This hints at the usefulness of SFA for autonomous learning on high-dimensional input streams . In fact , it was shown in [15] that important stimulus features such as object category , the position of objects , or their orientation can be easily extracted by supervised training with high precision from the slow features of a high-dimensional visual input stream . It should be noted that the SFA algorithm is only one particular implementation of learning based on slowness , and there have been various earlier approaches , e . g . , [16]–[19] . Slowness has previously been used in some hierarchical models as well [20]–[22] . Unsupervised learning based on the slowness principle ( i . e . , learning that exploits temporal continuity of real-world stimuli ) has recently attracted the attention of experimentalists [23] , [24] . It was shown in monkey experiments , that features in monkey infero temporal cortex are adapted in a way that is consistent with the slowness principle [23] . In this article , we propose a learning system where the state space representation is constituted autonomously by SFA . A subsequent neural circuit is then trained by a reward-based synaptic learning rule that is related to policy gradient methods or Q-learning in classical reinforcement learning . We apply this system to two closed-loop control tasks where the input to the system is high-dimensional raw pixel data and the output are motor commands . We thus show in this article for two control tasks on high-dimensional visual input streams that the representation of the SFA output is well suited to serve as a state-representation for reward-based learning in a subsequent neural circuit .
We tested this learning system on two different control tasks where an agent ( a fish ) navigates in a 2D environment with analog state- and action-space: a task similar to the Morris water-maze task [25] and a variable-targets task , see section “Tasks” . The state of the universe at time ( see below for details ) was used to render a 155 155 dimensional 2D visual scene that showed the agent ( a fish; for one of the tasks two fish-types with different visual appearance were used ) at a position and potentially other objects , see Figure 2 . This visual scene constituted the input to the learning system . These tasks are to be seen as generic control tasks of reasonable complexity . The bird's eye perspective used here is of course not realistic for animal agents . As demonstrated in [14] our model should also be able to deal with a first-person perspective , especially in the Morris water-maze . For the variable-targets task this would introduce some complications like the target not being in the field of view or being hidden behind the distractor . On the other hand it would simplify the task , since the agent would not need to know its own position and angle ( it could simply center its field of view on the target ) . For the training of the system , we distinguish two different phases . In a first phase the SFA network is trained . In this phase , the fish , the target , and the distractor are floating slowly over the 2D space of the environment . The type of fish is changed from time to time ( see section “Training stimuli of the hierarchical network” ) . In a second phase the control circuit is trained . This phase consists of several learning episodes , an episode being one trial to reach a defined target from the initial fish-position . An episode ends when the target is reached or when a maximum number of time-steps is exceeded . The hierarchical network described in the next section is based on the Slow Feature Analysis Algorithm ( SFA ) [26] , [27] . SFA solves the following learning task: Given a multidimensional input signal we want to find instantaneous scalar input-output functions that generate output signals that vary as slowly as possible but still carry significant information . To ensure the latter we require the output signals to be uncorrelated and have unit variance . In mathematical terms , this can be stated as follows: Optimization problem: Given a function space and an I-dimensional input signal find a set of real-valued input-output functions such that the output signals ( 1 ) under the constraints ( 2 ) ( 3 ) ( 4 ) with and indicating temporal averaging and the derivative of , respectively . Equation ( 1 ) introduces the -value , which is a measure of the temporal slowness ( or rather fastness ) of the signal . It is given by the mean square of the signal's temporal derivative , so that small -values indicate slowly varying signals . The constraints ( 2 ) and ( 3 ) avoid the trivial constant solution and constraint ( 4 ) ensures that different functions code for different aspects of the input . Because of constraint ( 4 ) the are also ordered according to their slowness , with having the smallest . It is important to note that although the objective is slowness , the functions are instantaneous functions of the input , so that slowness cannot be achieved by low-pass filtering . Slow output signals can only be obtained if the input signal contains slowly varying features that can be extracted instantaneously by the functions . Note also that for the same reason , once trained , the system works fast , not slowly . In the computationally relevant case where is finite-dimensional the solution to the optimization problem can be found by means of Slow Feature Analysis ( SFA ) [26] , [27] . This algorithm , which is based on an eigenvector approach , is guaranteed to find the global optimum . Biologically more plausible learning rules for the optimization problem exist [28] , [29] . The visual system is , to a first approximation , structured in a hierarchical fashion , first extracting local features which are then integrated to more and more global and abstract features . We apply SFA in a similar hierarchical manner to the raw visual input data . First , the slow features of small local image patches are extracted . The integration of spatially local information exploits the local correlation structure of visual data . A second layer extracts slow features of these features ( again integrating spatially local patches ) , and so on . Such hierarchical architecture is promising because SFA has been applied successfully to visual data in a hierarchical fashion previously [15] , [30] . A hierarchical organization also turns out to be crucial for the applicability of the approach for computational reasons . The application of non-linear SFA on the whole high-dimensional input would be computationally infeasible . Efficient use of resources is also an issue in biological neural circuits . It has been suggested that connectivity is the main constraint there [31] , [32] . Since a hierarchical organization requires nearly exclusively local communication , it avoids extensive connectivity . The hierarchical network consists of a converging hierarchy of layers of SFA nodes , and the network structure is identical to that used in [30] ( there this part of our model is also discussed in greater length ) . All required building blocks for the hierarchical network are available in the “Modular toolkit for Data Processing” ( MDP ) library [33] . We employed neural implementations of two reinforcement learning algorithms , one is based on Q-learning and one is a policy-gradient method . Neural versions of Q-learning have been used in various previous works on biological reward-based learning , see e . g . [34] , [35] . The popularity of Q-learning stems from the finding that the activity of dopaminergic neurons in the ventral tegmental area is related to the reward-prediction error [4] , [36] , [37] , a signal that is needed in Q-learning [35] . In Q-learning , decisions are based on a so-called Q-function that maps state-action pairs onto values that represent the current estimate of the expected total discounted reward given that action is executed at state . For a given state , the action with highest associated Q-value is preferred by the agent . However , to ensure exploration , a random action may be chosen with some probability . We implemented the neural version of Q-learning from [35] where the Q-function is represented by a small ensemble of neurons and parametrized by the connection weights from the inputs to these neurons . The system learns by adaptation of the Q-function via the network weights . In the implementation used in this article , this is achieved by a local synaptic learning rule at the synapses of the neurons in the neuron ensemble . The global signal that modulates local learning is the temporal difference error ( TD-error ) . We do not address in this article the question how this signal is computed by a neuronal network . Several possible mechanisms have been suggested in the literature [37]–[39] . The Q-function was represented by a set of linear neurons that receive information about the current state from the output of the SFA circuit . The output of neuron is hence given by . Each neuron has a dedicated preferred direction . The Q-value of a movement in direction for the given state is hence given by . The activities of these neurons imply a proposed action for the agent which is a movement in the direction given by the population vector . Here , is the angle of the vector ( 5 ) where the vector is the unit vector in direction . The Q-function is parametrized by the weight values and it is learned by adapting these weights according to the Q-learning algorithm ( see [35] ) : See Supporting Text S1 for parameter settings . The second learning algorithm employed was a policy gradient method . In this case , the action is directly given by the output of a neural network . Hence , the network ( which receives as input the state-representation from the SFA network ) represents a policy ( i . e . , a mapping from a state to an action ) . Most theoretical studies of such biologically plausible policy-gradient learning algorithms are based on point-neuron models where synaptic inputs are weighted by the synaptic efficacies to obtain the membrane voltage . The output of the neuron is then essentially obtained by the application of a nonlinear function to the membrane voltage . A particularly simple example of such a neuron model is a simple pseudo-linear rate-based model where a nonlinear activation function ( commonly sigmoidal ) is applied to the weighted sum of inputs : ( 6 ) Here , denotes the synaptic efficacy ( weight ) of synapse that projects from neuron to neuron , is a bias , and denotes some noise signal . We assume that a reward signal indicates the amount of reward that the system receives at time . Good actions will be rewarded , which will lead to weight changes that in turn make such actions more probable . Reinforcement learning demands exploration of the agent , i . e . , the agent has to explore new actions . Thus , any neural system that is subject to reward-based learning needs some kind of stochasticity for exploration . In neuron model ( 6 ) exploration is implemented via the noise term . Reward-based learning rules for this model can easily be obtained by changing the weights in the direction of the gradient of ( 7 ) where denotes the low-pass filtered version of with an exponential kernel , and is a small learning rate . In our simulations we used for the filtered reward . The update equations for the bias is analogous with . A single neuron of type ( 6 ) turns out to be too weak for some of the control tasks considered in this article . The standard way to increase the expressive power is to use networks of such neurons . The learning rule for the network is then unchanged , each neuron tries to optimize the reward independently from the others [40] , but see [41] . It can be shown that such a greedy strategy still performs gradient ascent on the reward signal . However , the time needed to converge to a good solution is often too long for practical applications as shown in Results . We therefore propose a learning rule that is based on a more complex neuron model with nonlinear dendritic interactions within neurons [42] and the possibility to adapt dendritic conductance properties [43] . In this model , the total somatic input to neuron is modeled as a noisy weighted linear sum of signals from dendritic branches ( 8 ) where describes the coupling strength between branch and the soma and is a bias . Again , models exploratory noise . At each time step , an independent sample from the zero mean distribution is drawn as the exploratory signal . In our simulations , is the uniform distribution over the interval . The output of neuron at time is modeled as a nonlinear function of the total somatic input: ( 9 ) Each dendritic branches itself sums weighted synaptic inputs followed by a sigmoidal nonlinearity ( 10 ) where denotes the synaptic weight from input to the dendritic branch of neuron . Update equations that perform gradient ascent on a reward-signal are derived in Supporting Text S2 . The derived update rules for the parameters are ( 11 ) ( 12 ) where and are small learning rates . The update rules can be extended to use eligibility traces that collect the information about recent pre-and postsynaptic states at the synapse in a single scalar value . In this way , previous states of the synapse can be incorporated in the weight change at time , which is driven by the momentary reward signal . In this article however , we rely on the update rules ( 11 ) and ( 12 ) without eligibility traces . See [41] for an alternative rule of similar flavor . In our simulations , we needed two control variables , one to control the speed of the agent and one for its angular velocity . Each control variable was computed by a single neuron of this type where each neuron had branches . The nonlinearity in the branches was the tangens hyperbolicus function . Also a logistic sigmoidal was tested which is a scaled version of the tangens hyperbolicus to the image set . Results were similar with a slight increase in learning time . The nonlinearity at the soma was the tangens hyperbolicus for the angular velocity and a logistic sigmoid for the speed . The noise signal was drawn independently for each neuron and at each time step from a uniform distribution in . Detailed parameter settings used for the simulations can be found in Supporting Text S1 . We tested the system on two different control tasks: a task similar to the Morris water-maze task and a variable-targets task .
We implemented this task with our learning system where the decision circuit consisted of the Q-learning circuit described above . In this task , the slowest components as extracted by the hierarchical SFA network were used by the subsequent decision network . The results of training are shown in Figure 5 . The performance of the system was measured by the time needed to reach the target ( escape latency ) . The system learns quite fast with convergence after about 40 training episodes . The results are comparable to previously obtained simulation results [34] , [35] , [44] that were based on a state representation by neurons with place-cell-like behavior . Figure 5B shows the direction the system chooses with high probability at various positions in the water maze ( navigation map ) after training . Using only the 16 slowest SFA components for reinforcement learning , the system has rapidly learned a near-optimal strategy in this task . This result shows that the use of SFA as preprocessing makes it possible to apply reinforcement learning to raw image data in the Morris water maze task . The Morris water maze task is relatively simple and does not provide rich visual input . We therefore tested the learning system on the variable-targets task described above , a control task where two types of fish navigate in a 2D environment . In the environment , two object positions were marked by a cross and a disk , and these positions were different in each learning episode . A target object was defined for each fish type and the task was to navigate the current fish to its target by controlling the forward speed and the change in movement direction ( angular velocity ) . The control of angular velocity , the arbitrary target position , and the dependence of the target object on the fish identity complicates the control task such that the Q-learning algorithm used in the water-maze task as well as a simple linear decision neuron like the one of equation ( 6 ) would not succeed in this task . We therefore trained the leaning system with the more powerful policy gradient algorithm described above on the slowest 32 components extracted by the hierarchical SFA network . In order to compute the SFA output fast , we had to perform the training of the control network in batches of 100 parallel traces in this task ( i . e . , 100 training episodes with different initial conditions are simulated in parallel with a given weight vector . After the simulation of a single time step in all 100 episodes , weight changes over these 100 traces are averaged and implemented . Then , the next time step in each of the 100 traces is simulated and weights are updated ) . When the agent in one of the traces arrived at the target , a new learning episode was initiated in this trace while other traces simply continued . As will be shown below , the training in batches has no significant influence on the learning dynamics . Results are shown in Figure 6A , B . The reward converges to a mean reward above which means that the agent nearly always takes the best step towards the target despite the high amount of noise in the control neurons . Figure 7 shows that the trajectories after training were very good . Interestingly , the network does not learn the optimal strategy with respect to the forward speed output . Although it would be beneficial to reduce the forward speed when the agent is directed away from the target , first rotate the agent , and only then move forward , the output of the speed neuron is nearly always close to the maximum value . A possible reason for this is that the agent is directed towards the target most of the time . Thus , the gain in reward is very small and a relatively small fraction of training examples demands low speed . We compared the results to a learning system with the same control circuit , but with SFA replaced by a vector which directly encoded the state-space in a straight-forward way . For this task with two fish identities and two objects , we encoded the state-space by a vector ( 17 ) where is the position of the agent , is its orientation , is its identity , and is the position of the object . Figure 6C , D shows the results when the control network was trained with identical parameters but with this state-vector as input . The Performance with the SFA network is comparable to the performance of the system with a highly informative and precise state encoding . For efficiency reasons , we had to perform the training of the control network in batches of 100 traces ( see above ) . Because no SFA is needed in the setup with the direct state-vector as input , we can compare learning performance of the control network to performance without batches . The result is shown in in Figure 6C , D ( gray dashed lines ) . The use of small batches does not influence the learning dynamics significantly . In the environment considered , movement is mirrored if the agent hits a boundary . Since this helps to avoid getting stuck in corners we performed control experiments where the movement in the direction of the boundary is simply cut off but no reflection happens ( i . e . , the dynamics of the position of the fish is given by and , compare to equations ( 14 ) , ( 15 ) ) . Results are shown in Figure S1 . As expected , the system starts with lower performance and convergence takes about twice as long compared to the environment with mirrored movements at boundaries . Interestingly , in this slightly more demanding environment , the SFA network is converging faster than the system with a highly informative and precise state encoding . In another series of experiments we tested how the performance depends on the number of outputs from the SFA network that are used as input for the reinforcement learning . Since the outputs of the SFA network are naturally ordered by their slowness one can pick only the first outputs and train the reinforcement learning network on those . For the variable-targets task we tested the performance for 16 , 22 , 28 , 32 , and 64 outputs . For 16 outputs the average reward value always stayed below and rose much slower than in the case of 32 outputs . For 28 outputs the performance was already very close to that of the 32 outputs . Going from 32 outputs to 64 did not change the average reward , but in the case of 64 outputs the trajectories of the agent occasionally showed some errors ( e . g . , the agent initially chose a wrong direction and took therefore longer to reach the target ) . We compared performance of the system to a system where the control network is a two-layer feed-forward network of simpler neurons without dendritic branches , see Equation ( 6 ) . We used two networks with identical architecture , one for each control variable . Each network consisted of 50 neurons in the first layer connected to one output neuron ( increasing the number of neurons in the first layer to 100 did not change the results ) . Every neuron in the first layer received input from all SFA outputs . The learning rates of all neurons were identical . See Supporting Text S1 for details on parameters and their determination . Results are shown in Figure S2 . The network of simple neurons can solve the problem in principle , but it converges much slower . We also compared performance of the system with SFA to systems where the dimensionality of the visual input was reduced by PCA . In one experiment the SFA nodes in the hierarchical network were simply replaced by PCA nodes . We then used 64 outputs from the network for the standard reinforcement learning training . As shown in Figure 8 the control network was hardly able to learn the control task . This is also evident in the test trajectories , which generally look erratic . In another experiment we used PCA on the whole images . Because of the high dimensionality we first had to downsample the image data by averaging over two by two pixels ( reducing the dimensionality by a factor of four ) before using linear PCA . The performance was very similar to the hierarchical PCA experiment ( the average reward hovered below ) . A direct analysis of the PCA output with linear regression [15] indicates that except for the agent identity , no important features such as position of the agent or the targets can be extracted in a linear way from the reduced state representation . For hierarchical SFA , such an extraction is often possible [15] . This hints at the possibility that the state representation given by PCA cannot be exploited by the control network because the implicit encoding of relevant variables is either too complex or too much important information has been discarded .
Several theoretical studies have investigated biologically plausible reward-based learning rules [46]–[55] . On the synaptic level , such rules are commonly of the reward-modulated Hebbian type , also called three-factor rules . In traditional Hebbian learning rules , changes of synaptic plasticity at time are based on the history of the presynaptic and the postsynaptic activity , such that the weight change of a synapse from a presynaptic neuron to a postsynaptic neuron is the product between some function of the presynaptic activity history and some function of the postsynaptic activity history . A third signal that models the local concentration of some neuromodulator which in turn signals some reward , is in many models modulating these Hebbian updates . Such update rules are either purely phenomenological [53] , [55] or derived from a reward-maximization principle [47]–[51] . From the viewpoint of classical reinforcement learning , the latter approach is related to policy-gradient methods . Since the learning algorithms in these previous works are based on simple neuron models , they are too weak for the variable-targets task considered in this article . The policy-gradient method used in this article extends the classical single-neuron based policy-gradient approach in the sense that it is based on a more expressive neuron model with nonlinear branches . In this model , both , synaptic weights and branch strengths are adapted through learning . Our approach is motivated by recent experimental findings where it has been shown that not only synaptic efficacies but also the strengths of individual dendritic branches are plastic [43] . Furthermore , it was shown that this type of plasticity is dependent on neuromodulatory signals . Our results ( compare Figure 6 to Figure S2 ) indicate that the neuron model with nonlinear branches can be trained much faster than networks of point-neuron models . This hints at a possible role of nonlinear branches in the context of reward-based learning . The Morris water-maze task has been modeled before . In [45] , a network of spiking neurons was trained on a relatively small discrete state-space that explicitly coded the current position of the agent on a two-dimensional grid . The authors used a neural implementation of temporal difference learning . In contrast to the algorithms used in this article , their approach demands a discrete state space . This algorithm is therefore not directly applicable to the continuous state-space representation that is achieved through SFA . In [34] and [44] the input to the reinforcement learning network was explicitly coded similar to the response of hippocampal place-cells . In [35] , the state-representation was also governed by place-cell-like response that were learned from the input data . This approach was however tailored to the problem at hand , whereas we claim that SFA can be used in a much broader application domain since it is not restricted to visual input . Furthermore , in this article SFA was not only used to extract position of an agent in space but also for position of other objects , for object identity , and for orientation . We thus claim that the learning architecture presented is very general only relying on temporal continuity of important state variables . Although the variable-targets task considered above is quite demanding , the learning system gets immediate feedback of its performance via the reward signal defined by equation ( 16 ) . By postulating such a reward signal one has to assume that some system can evaluate that “getting closer to the target” is good . Such prior knowledge could have been acquired by earlier learning or it could be encoded genetically . An example of a learning system that probably involves such a circuitry ( the critique ) is the song-learning system in the songbird . In this system , it is believed that a critique can evaluate similarity between the own song and a memory copy of a tutor song [56] . However , there is no evidence that such higher-level critique is involved for example in navigational learning of rodents . Instead , it is more natural to assume that an internal reward signal is produced for example when some food-reward is delivered to the animal . One experimental setup with sparse rewards is the Morris water maze task [25] considered above . In principle , this sparse reward situation could also be learned if the learning rules ( 11 ) , ( 12 ) are amended with eligibility traces [48] . However , the learning would probably take much longer . Given the high-dimensional visual encoding of the state-space accessible to the learning system , it is practically impossible that any direct reinforcement learning approach is able to solve the variable-targets task directly on the visually-induced state-space . Additionally , in order to scale down the visual input to viable sizes , a hierarchical approach is most promising . Here , hierarchical SFA is one of the few approaches that have been proven to work well . Linear unsupervised techniques such as principal component analysis ( PCA ) or independent component analysis ( ICA ) are less suited to be applied hierarchically . To understand the results , it is important to note that SFA is quite different from PCA or other more elaborate dimensionality reduction techniques [57] , [58] . Dimensionality reduction in general tries to produce a faithful low-dimensional representation of the data . The aim of SFA is not to produce a faithful representation in the sense that the original data can be reconstructed with small error . Instead , it tries to extract slow features by taking the temporal dimension of the data into account ( this dimension is not exploited by PCA ) and disregards many details of the input . Although it is in general not guaranteed that slowly varying features are also important for the control task , slowly varying features such as object identities and positions are important in many tasks . In fact , the removal of details may underlie the success of the generic architecture , since it allows the subsequent decision circuit to concentrate on a few important features of the input . This may also explain the failure of PCA . The encoding of the visual input produced by PCA can be used to reconstruct a “blurred” version of the input image . However , it is very hard to extract from this information the relevant state variables such as object identity or position . But this information can easily be extracted from the SFA output , see [15] . We compared the preprocessing with SFA to PCA preprocessing but not to more elaborate techniques [57] , [58] since the focus of this paper is on simple techniques for which some biological evidence exists . Another candidate for sensory preprocessing instead of SFA is ICA . However , ICA does not provide a natural ordering of extracted components . It is thus not clear which components to disregard in order to reduce the dimensionality of the sensory input stream . One interesting possibility would be to order the ICA components by kurtosis in order to extract those components which are most non-Gaussian . Another interesting possibility not pursued in this paper would be to sparsify the SFA output by ICA . This has led to place-cell like behavior in [14] and might be beneficial for subsequent reward-based learning . Information bottleneck optimization ( IB ) is another candidate learning mechanism for cortical feature extraction . However , IB is not unsupervised , it needs a relevance signal . It would be interesting to investigate whether a useful relevance signal could be constructed for example from the reward signal . Finally , the problem of state space reduction has also been considered in the reinforcement learning literature . There , the main approach is either to reduce the size of a discrete state space or to discretize a continuous state-space [59] , [60] . In contrast , SFA preserves the continuous nature of the state-space by representing it with a few highly informative continuous variables . This circumvents many problems of state-space discretization such as the question of state-space granularity . Thus , there are multiple benefits of SFA in the problem studied: It can be trained in a fully unsupervised manner ( as compared to IB ) . By taking the temporal dimension into account , it is able to compress the state-space significantly without the need to discretize the continuous state-space ( as compared to [59] , [60] ) . It provides a highly abstract representation that can be utilized by simple subsequent reward-based learning ( compare to the discussion of PCA ) . The possibility to apply SFA in a hierarchical fashion renders it computationally efficient even on high-dimensional input streams , both in conventional computers and in biological neural circuits where it allows for mainly local communication and thus avoids extensive connectivity [31] , [32] . The natural ordering of features based on their slowness implies a simple criterion on the basis of which information can be discarded in each node of the hierarchical network ( compare to ICA ) , resulting in a significant reduction of information that has to be processed by higher-level circuits . Finally , SFA is relatively simple , its complexity is comparable to PCA and it is considerably simpler than other approaches for state-space reduction [57]–[60] . Accordingly , biologically plausible implementations of SFA exist [28] , [29] . Together with the fact that experimental evidence for slowness learning exists in the visual system [23] , this renders SFA an important candidate mechanism for unsupervised feature extraction in sensory cortex . In this article , we provided a proof of concept that a learning system with an unsupervised preprocessing and subsequent simple biologically realistic reward-based learning can learn quite complex control tasks on high-dimension visual input streams without the need for hand-design of a reduced state-space . We applied the proposed learning system to two control tasks . In the Morris water maze task , we showed that the system can find an optimal strategy in a number of learning episodes that is comparable to experimental results with rats [25] . The application of the learning system to the variable targets task shows that also much more complex tasks with rich visual inputs can be solved by the system . We propose in this article that slowness-learning in combination with reward-based learning may provide a generic ( although not exclusive ) principle for behavioral learning in the brain . This hypothesis predicts that slowness learning should be a major unsupervised learning mechanism in sensory cortices of any modality . Currently , such evidence exists for the visual pathway only [23] . We showed that learning performance of the system in this task is comparable to a system where the state-representation extracted by SFA is replaced by a highly compressed and precise hand-crafted state-space . Finally , our simulation results suggest that performance of the system is quite insensitive to the number of SFA components that is chosen for further processing by the reinforcement learning network as long as enough informative features are chosen . Altogether this study provides , on the one hand , further support that slowness learning could be one important ( but not necessarily exclusive ) unsupervised learning principle utilized in the brain to form efficient state representations of the environment . On the other hand , this work shows that autonomous learning of state-representations with SFA should be further pursued in the search for autonomous learning systems that do not - or much less - have to rely on expensive tuning by human experts . | Humans and animals are able to learn complex behaviors based on a massive stream of sensory information from different modalities . Early animal studies have identified learning mechanisms that are based on reward and punishment such that animals tend to avoid actions that lead to punishment whereas rewarded actions are reinforced . It is an open question how sensory information is processed by the brain in order to learn and perform rewarding behaviors . In this article , we propose a learning system that combines the autonomous extraction of important information from the sensory input with reward-based learning . The extraction of salient information is learned by exploiting the temporal continuity of real-world stimuli . A subsequent neural circuit then learns rewarding behaviors based on this representation of the sensory input . We demonstrate in two control tasks that this system is capable of learning complex behaviors on raw visual input . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"neuroscience/behavioral",
"neuroscience",
"neuroscience/natural",
"and",
"synthetic",
"vision",
"neuroscience/theoretical",
"neuroscience",
"computational",
"biology/computational",
"neuroscience"
] | 2010 | Reinforcement Learning on Slow Features of High-Dimensional Input Streams |
Gene-expression deconvolution is used to quantify different types of cells in a mixed population . It provides a highly promising solution to rapidly characterize the tumor-infiltrating immune landscape and identify cold cancers . However , a major challenge is that gene-expression data are frequently contaminated by many outliers that decrease the estimation accuracy . Thus , it is imperative to develop a robust deconvolution method that automatically decontaminates data by reliably detecting and removing outliers . We developed a new machine learning tool , Fast And Robust DEconvolution of Expression Profiles ( FARDEEP ) , to enumerate immune cell subsets from whole tumor tissue samples . To reduce noise in the tumor gene expression datasets , FARDEEP utilizes an adaptive least trimmed square to automatically detect and remove outliers before estimating the cell compositions . We show that FARDEEP is less susceptible to outliers and returns a better estimation of coefficients than the existing methods with both numerical simulations and real datasets . FARDEEP provides an estimate related to the absolute quantity of each immune cell subset in addition to relative percentages . Hence , FARDEEP represents a novel robust algorithm to complement the existing toolkit for the characterization of tissue-infiltrating immune cell landscape . The source code for FARDEEP is implemented in R and available for download at https://github . com/YuningHao/FARDEEP . git .
Immune checkpoint blockade has revolutionized the rational design of neoadjuvant cancer therapies . Compelling evidence suggests that a favorable tumor immune microenvironment underpins better clinical responses to radiotherapy , chemotherapy , and immunotherapy [1–3] . Immunohistochemistry ( IHC ) -based immunoscores , which quantify the number of CD8+ cytotoxic T lymphocytes and CD45RO+ memory T cells , show better prognostic potential than conventional pathological methods in colon cancer patients [4 , 5] . Hence , harnessing the composition of intra-tumoral immune cell infiltration is a highly promising approach to stratify tumors [6–11] . The current IHC immunoscoring approach has two limitations . First , the interpretation of immune cell subsets varies among pathologists and institutions , thus lacking a consistent standard for the scoring practice . Second , only a limited number of biomarkers can be assessed simultaneously , which prevents a comprehensive annotation of the immune contexture in the tumor microenvironment ( TME ) . Hence , robust methods for genome data-informed cell type quantitation are in urgent need . Immunogenomics presents an unprecedented opportunity to resolve the intra-tumoral immune landscape . Cell type deconvolution using leukocyte signature gene expression profiling is a highly promising approach to estimate the global immune cell composition from whole tumor gene expression data [12–17] . However , a significant technical obstacle is that the efficacy and accuracy of gene expression deconvolution are limited by the large number of outliers , which are frequently observed in tumor gene expression datasets [18] . The first step towards enhancing the overall gene deconvolution algorithms is to improve methods for outliers identification and processing . Those outliers include genes with abnormal expression value which may be caused by measurement error , environmental effect , expression from non-immune cells , or natural fluctuations in certain type of immune cells . Notably , the current immune deconvolution gene signature matrix relies on the profiling of differentially expressed genes among different immune subsets . Frequent contamination of transcripts reading from cancer cells may significantly bias the algorithms . In this study , we report a novel FAst and Robust DEconvolution of Expression Profiles ( FARDEEP ) method that significantly improves the estimation of coefficients . Let yi be the observed expression value for the ith gene; xi , a p-dimensional vector , be the expected expression of the ith gene for the p different cell types; and X = [x1 , … , xn]′ be the signature matrix . The gene-expression deconvolution problem can be formulated as follows , y i = x i ′ β + ε i , ( 0 . 1 ) where β ∈ R p is an unknown parameter corresponding to the compositions of p cell types , and εi is a noise term with a mean of 0 . Several methods were proposed to solve this deconvolution problem . To enforce the non-negativity of β in ( 0 . 1 ) , several algorithms , such as the Non-Negative Least Square ( NNLS ) , Non-negative Maximum Likelihood ( NNML ) frameworks and the perturbation model ( PERT ) were developed . They all rely on the signature matrix ( X ) derived from Microarray experiments [14 , 19–24] . To extend this work to RNA-seq data , Finotello et al . [14] proposed a constraint linear model with a signature matrix derived from RNA-seq data . Additionally , the gene expression of each cell may vary depending on its microenvironment and other factors , which will lead to a biased estimation . To address this issue , Microarray Microdissection with Analysis of Differences ( MMAD ) incorporates the concept of the effective RNA fraction and estimates coefficients using a maximum likelihood approach [25] . To further adapt deconvolution to high-dimensional settings , Altboum et al . [26] proposed a penalized regression framework , Digital Cell Quantifier ( DCQ ) , to encourage sparsity for the estimated β using the elastic net [27] . Cell-type identification by estimating relative subsets of RNA transcripts ( CIBERSORT ) uses ν-support vector regression ( ν-SVR ) to enhance the robustness of gene expression deconvolution . CIBERSORT performs a regression by finding a hyperplane that fits as many data points as possible within a tube whose vertical length is a constant ε [12] . The ε-tube provides a region in which estimation errors are ignored . This model does not include an intercept to capture contributions of other contents . Additionally , to increase the computational efficiency , CIBERSORT applies Z-normalization to the data before fitting the regression , which may introduce estimation bias . Based on the CIBERSORT framework , several extensions have been proposed to overcome limitations such as platform inconsistency between signature and mixture matrices and low estimation accuracy for γδ T cell [15–17] . However , the quantitative information of cell proportions of these two approaches is built on CIBERSORT whose performance may be challenged by frequent outliers in whole tumor tissue transcriptomes . To reduce the dependence on the signature matrix , xCell utilizes the concept of single-sample gene set enrichment analysis ( ssGSEA ) to calculate an immune cell score which could predict the enrichment of immune cells [13] . Despite its robustness , xCell relies much on the ranking of gene expression value which leads to suboptimal solution for the estimation accuracy . Overall , a robust method that determines both the distribution and absolute volume of tumor-infiltrating lymphocytes ( TILs ) will further improve the current gene deconvolution pipeline . To handle the heavily contaminated gene expression data and provide absolute cell abundance estimation , we developed a robust method based on the Least Trimmed Square ( LTS ) framework [28 , 29] . LTS finds h observations with smallest residuals , and the estimator β ^ is the least squares fit over these h observations . LTS is an NP-hard problem , and Rousseeuw and Driessen [30] proposed a stochastic FAST-LTS algorithm . Nevertheless , it may give a suboptimal fitting result and get much slower when the sample size and dimension of variables become larger and higher since its accuracy relies on the initial random h-subsets and the number of initial subsets . When n is the sample size and p is the number of coefficients , h is suggested to be the smallest integer that is not less than ( n + p + 1 ) /2 to remove as many outliers as possible while keeping an unbias result . Using the information of only half of the data reduces the power of the estimator because the amount of outliers in the real case cannot be presumed and can be small . Xu et al . [31] proposed an adaptive least trimmed square which is not limited to the randomness of initial subset but only applied the binary dataset . In this study , we extend the adaptive least trimmed square to introduce a model-free method , which could find the number of outliers automatically based on LTS . FARDEEP provides a flexible framework which is suitable for both Microarray and RNA-seq data using LM22 and Immunostate signature matrices respectively . As evidence of high fidelity and robustness , FARDEEP exhibits superior performance in simulated and real-world datasets .
The usual linear deconvolution model can be expressed as below , y = X β + ε , where y ∈ R n is the observed expression data for n immune subset signature genes , X ∈ R n × p denotes a mean gene expression signature matrix for p different cell types , β ∈ R p contains each unknown cell type abundance , and ε ∈ R n is a vector of random errors with zero mean and variance of σ2I . To incorporate outliers , we propose the following model y = X β + τ + ε , ( 0 . 2 ) where parameter τ = ( τ1 , … , τn ) ′ is a sparse vector in R n with τi ≠ 0 indicating the ith gene is an outlier . Under the formulation of ( 0 . 2 ) , let β ^ ols = ( X ⊤ X ) − 1 X ⊤ y be the Ordinary Least Square ( OLS ) estimate and H = X ( X⊤X ) −1X⊤ be the projection matrix . The residuals r = ( r1 , … , rn ) using OLS could be formulated as r = y − X β ^ ols = ( I − H ) τ + ( I − H ) ϵ . ( 0 . 3 ) with mean of ( I − H ) τ and variance of σ2 ( I − H ) . From ( 0 . 3 ) , the residuals , ri with the corresponding τi ≠ 0 , would deviate from zero , which suggests that the set of outliers can be identified through thresholding as follows E = { i : | r i | > k × r med } , ( 0 . 4 ) where E is the set of detected outliers , k is a tuning parameter controlling the sensitivity of the model , and rmed is the median of { | r | i } i = 1 n . We denote the number of elements in set E as |E| and let N be the number of true outliers in the data . First , we can use least squares and formula ( 0 . 4 ) to obtain a rough estimate of E denoted as E ^ . Let the cardinality of E ^ be N ¯ . Since the model at this moment is inaccurate with contamination of outliers , N ¯ is an overestimation of N which can be used to get an underestimate via N _ = α 1 N ¯ with α1 ∈ ( 0 , 1 ) . With N _ , we can then update the least square fitting after removing the N _ samples with the largest absolute value of residuals and obtain an improved estimate of E and the corresponding N ¯ . We can improve the model by repeating the procedure , but we need to increase the underestimate of outliers , N _ , by a factor of α2 with α2 > 1 for each iteration to force the convergence between N ¯ and N _ . In sum , we initialize our algorithm by setting β ^ ( 0 ) = ( X ⊤ X ) − 1 X ⊤ y , r ( 0 ) = y − X β ^ ( 0 ) , which is the OLS solution . For the jth iteration , where j ≥ 1 , we update N ¯ ( j ) by N ¯ ( j ) = { | { i : | r i ( j − 1 ) | > r med ( j − 1 ) } | , j = 1 , min ( | { i : | r i ( j − 1 ) | > k · r med ( j − 1 ) } | , N ¯ ( j − 1 ) ) , j ≥ 2 . ( 0 . 5 ) where the min ( ⋅ , ⋅ ) operator guarantees that N ¯ ( j ) , an overestimation of N , is non-increasing . Similarly , we update N _ ( j ) through N _ ( j ) = { ⌈ α 1 N ¯ ( j ) ⌉ , j = 1 , min { ⌈ α 2 N _ ( j − 1 ) ⌉ , N ¯ ( j ) } , j ≥ 2 , ( 0 . 6 ) where ⌈x⌉ means the ceiling of x ∈ R , α1 ∈ ( 0 , 1 ) is used to obtain a lower bound for N in the first step , α2 > 1 guarantees the monotonicity of N _ ( j ) , and the min ( ⋅ , ⋅ ) operator guarantees N _ ( j ) is smaller than N ¯ ( j ) . Then we update β ^ and r after removing N _ ( j ) outliers by β ^ ( j ) = ( X ( j ) ⊤ X ( j ) ) − 1 X ( j ) ⊤ y ( j ) , r ( j ) = y − X β ^ ( j ) . We repeat this procedure until N _ and N ¯ converge . Hence , we hereby report a novel approach , coined as adaptive Least Trimmed Square ( aLTS ) , to automatically detect and remove contaminating outliers . Our aLTS is an extension of the iterative LTS algorithm proposed by Xu et al . [31] which is designed for binary output such as the comparison between two images or videos . Because the abundance of cell types are always non-negative , we replaced the OLS regression in the aLTS procedure with non-negative least square regression ( NNLS ) . By applying the modified aLTS to the deconvolution model ( 0 . 2 ) and solving the following problem , β ^ = argmin β‖ y − X β‖2 2 , subjecttoβ ≥ 0 using Lawson-Hanson algorithm [19] , we developed a robust tool , FARDEEP , for cellular deconvolution summarized in Algorithm 1 . One unique advantage of FARDEEP is that it is fast and guarantees to converge within finite steps , which is summarized in the following theorem . Algorithm 1 FAst and Robust DEconvolution of Expression Profiles Input: k > 0 , 0 < α1 < 1 , α2 > 1 , y , X Initialization: solving the following NNLS problem β ^ ( 0 ) = argmin β‖ y − X β ‖2 2 , subject to β ≥ 0 ; r ( 0 ) = y − X β ^ ( 0 ) . 1: compute N ¯ ( 1 ) and N _ ( 1 ) using ( 0 . 5 ) and ( 0 . 6 ) ; 2: solving the NNLS problem after removing N _ ( 1 ) genes with largest residuals , and update β ^ ( 1 ) , r ( 1 ) . 3: repeat 4: compute N ¯ ( j ) and N _ ( j ) using ( 0 . 5 ) and ( 0 . 6 ) for j ≥ 2; 5: solving the NNLS problem after removing N _ ( j ) genes with largest residuals , and update β ^ ( j ) , r ( j ) ; 6: until N ¯ = N _ . Output: Coefficients β ^ , Number of outliers N ^ , Index of outliers Theorem 1 Algorithm 1 ( FARDEEP ) stops in no more than j* steps , where j * = ⌊ − log α 1 log α 2 ⌋ + 2 . Here ⌊⋅⌋ is the largest integer that is less than or equal to x . Proof . It follows from the fact that the sequence { N ¯ ( j ) } is non-increasing , and { N _ ( j ) } is a geometrically increasing sequence that is bounded by the smallest component of { N ¯ ( j ) } . Specifically , assume that j* steps have been taken in FARDEEP , then j has approached j* − 1 , and N _ ( j ) ≥ α 2 N _ ( j − 1 ) for 0 ≤ j ≤ j* − 1 , so N ¯ ( 0 ) ≥ N ¯ ( j * − 2 ) ≥ N _ ( j * − 2 ) ≥ α 2 j * − 2 N _ 0 ≥ α 2 j * − 2 α 1 N ¯ 0 . which leads to the result . The β ^ from FARDEEP , denoted as TIL subset score , is the direct estimate of the linear model without any normalization and hence reflects the absolute abundance of TILs . In addition , we can derive the relative TILs abundance from the TIL subset scores through β ˜ j = β j ^ ∑ k = 1 p β j ^ , ( 0 . 7 ) where β ^ j is the jth TIL subset score . In practice , the TIL subset score provides important information of patient’s tumor-infiltrating immune landscape , and we have included a discussion in S2 Text . There are three tuning parameters k , α1 , and α2 in FARDEEP . Since α1 is only used in the first iteration , a relatively small α1 is preferred to ensure that FARDEEP does not remove too many outliers at the first step . In practice , FARDEEP is not sensitive to different values of α1 , and α2 , so we set them to 0 . 1 and 1 . 5 respectively by default . However , k controls the number of outliers in each iteration and is critical for the performance of FARDEEP . Thus , we tune k on a case-by-case basis for each sample to preserve meaningful fluctuations of gene expression levels . Effects for different tuning parameters are shown in S1 Table . Since the test group may contain outliers that influence the accuracy of the tuning result , cross-validation is not advised . Instead , we applied the Bayesian Information Criterion ( BIC ) and assume that the errors follow a log-normal distribution instead of a normal distribution among gene expression datasets as suggest by Beal [32] . We define the modified BIC referring to the setting of She and Owen [33]: BIC * ( k ) = mlog ∑ i = 1 n 1 { i ∉ E ^ }log 2 ( y i − y ^ i ) 2 m + b ( log ( m ) + 1 ) , ( 0 . 8 ) where E ^ being the set of detected outliers , b is number of parameters and equals N ^ + p + 1 with N ^ = | E ^ | being the number of outliers , and m equals n − N ^ . Then , we choose the value of k associated with the smallest BIC* .
To test the robustness of FARDEEP under different error conditions , we simulated three datasets refer to the setting in [33 , 34] with normally distributed errors , heavy tailed errors . The observations were generated based on the linear regression model ( 0 . 2 ) . The predictor matrix is X = ( x1 , … , xn ) ′ = UΣ1/2 , where U i j ∼ U ( 0 , 20 ) and Σ i j = ρ I { i ≠ j } with ρ = 0 . 5 . Consider the proportion of outliers f ∈ {5% , 10% , 20% , 30%} , sample size n = 500 , and number of predictors p = 20 , we added random errors and outliers to the simulated data as follows: Random errors: we generated the random error vector from i ) standard normal distribution , ii ) t-distribution with 3 degrees of freedom . Vertical outliers: we generated a n dimensional zero vector τ and randomly selected nf elements in τ to be the outliers generated from a non-central t-distribution with 1 degree of freedom and a non-centrality parameter of 30 . Leverage points: we took 20% of the contaminated data as leverage points , that is , replacing the corresponding predictors by the samples from N ( 2 max ( X ) , 1 ) . The coefficients βj were sampled from U ( 0 , 1 ) , where j = 1 , … , p . Based on the framework above , the dependent variable could be obtained by y = X β + τ + ε . We simulated each model 50 times . As shown in Figs 1 and 2 , FARDEEP outperforms other methods , evidenced by the SSE , R2 and R values . To check FARDEEP’s accuracy of outlier detection , we simulated {5%;10%;20%;30%} outliers using the same method as above for a model with both normally distributed and heavy-tailed noise . As shown in Table 1 , the tuning parameter k decreases when the amount of outliers becomes larger , and the true positive rates always stay around 1 , indicating that the tunning of k is highly effective . In the supplementary material S3 Text , we also included another outlier construction scheme with X related outliers and a simulation setting with correlated responses . In both scenarios , FARDEEP dominates other methods in terms of SSE , R2 and R values . Following the similar procedure as in Newman et al . , we randomly generated the abundance of different cells from interval [0 , 1] [12] . Notably , the sum of cell abundance is not necessarily 1 . The measurement errors were sampled from 2 N ( 0 , ( 0 . 1log 2 ( s ) ) 2 ) . To incorporate outliers , we randomly selected i/50 of the data and replaced them with data drawn from 2 N ( 10 , ( 0 . 3log 2 ( s ) ) 2 ) where i = 1 , 2 , … , 25 and s is the standard deviation of original mixtures . We repeated the procedure nine times and estimated the cell type abundance using FARDEEP , CIBERSORT ( without converting to percentage ) , NNLS , PERT , and DCQ . As shown in S2 Table , we found that the SSE range for FARDEEP is 1 . 51 × 10−7 to 1 . 47 × 10−4 , R2 and R keeps being 1 regardless of the number of outliers , while Other methods show significantly larger SSE and smaller R2 , R . We used the cell line dataset GSE11103 generated by Abbas et al . [35] that contains gene expression profiles of four immune cell lines ( Jurkat , IM-9 , Raji , and THP-1 ) and four mixtures ( MixA , MixB , MixC , and MixD ) with various ratios of cells . Before analysis , we quantile normalized the mixture data for 54675 probesets and downloaded the immune gene signature matrix with 584 probesets from CIBERSORT website . Then , we applied five deconvolution methods , including FARDEEP , CIBERSORT ( without converting to percentage ) , DCQ , NNLS , and PERT , to calculate the sum of squared errors of the estimated abundance of the four immune cell lines . We also compared with CIBERSORT absolute mode , which is a beta version in CIBERSORT website ( S1 Fig ) . Since the CIBERSORT absolute mode is a beta version and leads to suboptimal results compared with CIBERSORT , we only focused on the comparisons with CIBERSORT . We show that FARDEEP gives the smallest SSE and the largest R2 , which indicates the most accurate result ( Fig 3 ) . Both CIBERSORT and FARDEEP are robust deconvolution methods and show advantages in the above datasets , we next sought to compare their performances on mixtures with unknown content . We followed the simulation setting proposed by Newman et al . [12] and downloaded the signature gene file from CIBERSORT website . The mixture file was constructed from the four immune cell lines data , as mentioned in the previous section , and a colon cancer cell line HCT116 ( average of GSM269529 and GSM269530 in GSE10650 ) . Cancer cells were mingled into immune cells at different ratios {0% , 30% , 60% , 90%} . Noise was added by sampling from the distribution 2 N ( 0 , ( flog 2 ( s ) ) 2 ) , in which f ∈ {0% , 30% , 60% , 90%} and s is the standard deviation of original mixtures . By applying FARDEEP and CIBERSORT ( without converting to percentage ) on 64 mixtures , we found that FARDEEP remains an accurate estimation , while the tumor contents skew the results of CIBERSORT with larger deviation from the ground truth ( Fig 4 ) . To evaluate the performance of FARDEEP in immune-cell-rich settings that are less affected by outliers , we downloaded and analyzed two gene expression datasets ( GSE65135 [12] and GSE20300 [36] ) generated from the Affymetrix Microarray , which is the same platform used to generate the signature matrix LM22 . The GSE65135 dataset consists of ( i ) surgical lymph node biopsies of 14 follicular lymphoma patients and ( ii ) purified B and T cells from the tonsils of 5 healthy controls , and the GSE20300 dataset includes 24 blood samples from pediatric renal transplant patients . Flow cytometry or coulter counter data in these studies , which are presented in relative scales , are treated as ground truth . Thus , we normalized the estimated parameters of each method to the sum of 1 before comparison . As shown in Fig 5A and 5B for case ( i ) of GSE65135 and Fig 5D and 5E for GSE20300 , FARDEEP outperformed CIBERSORT in terms of R2 , R and SSE , which is consistent with our findings with simulated datasets . For case ( ii ) of GSE65136 , we estimated the immune cell composition for purified B and T cells with purity level exceeding 95% and 98% , respectively . For purified B cells , CIBERSORT tends to return non-zero estimates for T cell and a large proportion of other cell types , while FARDEEP gave almost all zero estimates for T cell and on average reduced the estimation error by 61% . Similarly , for the purified T cell , although CIBERSORT had a better performance compared to purified B cell , FARDEEP still significantly improves the estimation accuracy by reducing on average 48% of the estimation error ( Fig 5C ) . Furthermore , as shown in S4 Table , FARDEEP detected gene CD79A and BCL2A1 as outliers for most samples in case ( i ) of GSE65135 . These two genes are known to have high expression levels in follicular lymphoma ( B-cell lymphoma ) cells [37] . Overall , even in specimens that are rich in immune cells without contamination by non-hematopoietic malignancy , FARDEEP still outperforms CIBERSORT in immune cell deconvolution . In addition to effectively handling Microarray data , FARDEEP can also deconvolve TILs using RNA-seq data when we replace the signature matrix LM22 with quanTIseq , a signature matrix generated from RNA-seq data containing ten different immune cell types [14] . We applied CIBERSORT and FARDEEP using signature matrix quanTIseq to peripheral blood mononuclear cell ( PBMC ) mixtures ( GSE64655 ) generated by Hoek et al . [38] , and lymph node bulk samples of 4 melanoma patients from GSE93722 [39] . Flow cytometry data in these studies are on a relative scale and are treated as ground truth . We normalized the estimated parameters of each method to a relative scale using ( 0 . 7 ) before comparison . The RNA-seq data are usually less noisy compared to Microarray , and PBMC datasets are usually clean with less unknown contents . Therefore , we expect FARDEEP and CIBERSORT will return comparable results , which is the case in Fig 6A and 6B . However , when dealing with noisier data containing more outliers such as lymph node bulk samples , FARDEEP obtained larger advantage over CIBERSORT as shown in Fig 6C and 6D . TME of solid carcinomas are different from a lymph node biopsy or peripheral blood , and the highly variable gene expression in cancer cells challenges the accuracy of immune cell deconvolution . It is well-established that immune infiltration profile serves as a promising prognosticator [4 , 5] . Hence , we next utilized survival and gene expression data of ovarian cancer ( OV ) and lung squamous cell carcinoma ( LUSC ) from The Cancer Genome Atlas ( TCGA ) database to assess the prognostic relevance of different deconvolution methods . These two datasets were chosen because only LM22 not the RNA-seq based signature matrix quanTIseq includes γδ T cells , and OV and LUSC from TCGA datasets are the only two cancer types with Affymetrix microarray data . Using gene expression data ( n = 514 for OV and n = 133 for LUSC ) , we estimated the immunoscore using ESTIMATE proposed by yoshihara et al . [40] , TILs proportion using CIBERSORT , as well as TILs subset scores using CIBERSORT ( without converting to percentage ) and FARDEEP . Cold tumors typically harbor lower numbers of CD8+ T cells , γδ T cells , M1-like macrophages , and NK cells [11 , 41–43] . Thus , we calculated an anti-tumor immune subsets score by the summation of CD8+ T cells , γδ T cells , M1-macrophages , and NK cells . Then , we partitioned the patients into two groups with equal size using the median of either the immunoscore ( ESTIMATE ) or anti-tumor immune subsets score ( CIBERSORT and FARDEEP ) . We compared the survival curves between the two groups . As shown in Fig 7 , FARDEEP most effectively separates patients into high- and low- risk groups with the smallest p-value ( p-value = 0 . 0065 and 0 . 059 for OV and LUSC respectively ) . Recently , CIBERSORT website supports a beta-version of an absolute mode for cell deconvolution . We also included CIBERSORT absolute mode in this survival analysis and showed that it returned a better result ( p-value = 0 . 037 ) compared to the relative mode in the OV dataset . FARDEEP shows a stronger performance with a smaller p-value under this setting ( S2 Fig ) . These results demonstrated that the TIL subset scores could provide additional clinical-relevant information compared to the relative abundance . In addition , we expected the summation of these TIL subset scores would negatively correlate with tumor purity . To prove this hypothesis , we calculated the summation of 22 TIL subset scores for both OV and LUSC datasets and correlated them with the tumor purity estimated from consensus measurement of purity estimations ( CPE ) [44] . Even without taking account of stromal cells , as shown in S3 Fig . the summation of TIL subset scores is negatively correlated with tumor purity . Next , we sought to investigate whether outlier removal reduces contamination by transcripts from cancer cells . We first identified those top outlier-genes , which were consistently removed by FARDEEP in the OV dataset and obtained the average expression values of those outlier-genes from OV cell lines in GSE32474 [45] . As shown in S3 Table , most of these outlier-genes have high expression in cancer cell lines . For example , CXCL10 gene encodes an important chemokine to recruit CD8+ T cells and is also highly expressed in ovarian cancer cells . Thus , although some genes in LM22 may play a role in immune cells , they may be also highly expressed and variable among cancer cells . Such cross-contamination likely skews immune deconvolution analysis . As shown in S3 Table , FARDEEP successfully detected and removed those genes , leading to a more robust and accurate deconvolution analysis .
The cancer immune microenvironment has emerged as a critical prognostic dimension that modulates patient responses to neoadjuvant therapy . However , the current clinical TNM staging system does not have a consistent method to stratify cancers based on their immunogenicity . The recent study shows that the RNA-seq datasets of whole tumors contain valuable prognostic information to assess the cancer-immunity interactions [12 , 46] . But the current methods to extract immune signatures are susceptible to the frequent outliers in the datasets , leading to less effective identification of cold tumors . Based on support vector regression , CIBERSORT is one of the most popular robust deconvolution methods . However , this model does not include an intercept to capture possible contribution from other cell types and performs a z-normalization to the data before fitting the regression model , which introduces biases into the output . Discussion of the effect of Z-score normalization for CIBERSORT is included in S1 Text . In this study , we developed a new machine learning tool , FARDEEP , to streamline the removal of outliers and increase the robustness of gene-expression profile deconvolution . Robustness is an indispensable feature to solve a problem of deconvolution because gene expression data are frequently contaminated by a large amount of outliers . FARDEEP solves the deconvolution problem in a robust way because this tool evaluates all outliers across the datasets and then examines the true immune gene signature using non-negative regression . This feature is especially useful to analyze tumors with significant non-hematopoietic tumor components . Interestingly , although FARDEEP and the current robust methods can both tackle immune-cell-rich specimens such as lymph node and PBMCs , FARDEEP exhibits improved prognostic potential when dealing more complex datasets with significant carcinoma cell content . Although FARDEEP provides a robust computational algorithm to better solve the gene-expression deconvolution problem with noisy datasets , its performance and application rely on the choice of the signature matrix . To avoid estimation bias , it is important to choose the signature matrix derived from the same platform as the mixture matrix . For example , if dealing with gene expression data measured by Affymetrix HGU133A , we should use LM22 , but if dealing with RNA-seq data , the signature matrix quanTIseq is preferred . Overall , here we show that FARDEEP is a powerful and rapid machine learning tool that outperforms existing robust methods for gene deconvolution in datasets with significant heavy-tailed noise . FARDEEP provides a new technology to interrogate cancer immunogenomics and more accurately map the immune landscape of solid tumors . | Rapidly emerging evidence suggests that the tumor immune microenvironment not only predisposes cancer patients to diverse treatment outcomes but also represents a promising source of biomarkers for better patient stratification . Different from the immunohistochemistry-based scoring practice , which focuses on a few selected marker proteins , immune deconvolution pipelines inform a previously untapped method to comprehensively reveal the tumor-infiltrating immune landscape . Recognizing the numerous strengths of existing immune deconvolution algorithms , here we show data outliers , which are inevitable in whole tissue sequencing data sets , substantially skew estimation results . Moreover , an estimate related to the absolute amount of each immune subset offers valuable insight into the nature of the host response in addition to percentage information alone . Thus , we engineered a new immune deconvolution pipeline , coined as Fast and Robust Deconvolution of Expression Profiles ( FARDEEP ) , to automatically detect and remove outliers prior feeding data into the deconvolution algorithm and to provide estimates related to the absolute quantity of each immune subset . Utilizing both synthetic and real data sets , we found that FARDEEP returns superior coefficients and offers a robust tool to reveal the immune landscape of human cancers . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"engineering",
"and",
"technology",
"applied",
"mathematics",
"immunology",
"pert",
"simulation",
"and",
"modeling",
"algorithms",
"probability",
"distribution",
"mathematics",
"lymph",
"nodes... | 2019 | Fast and robust deconvolution of tumor infiltrating lymphocyte from expression profiles using least trimmed squares |
Genomic imprinting is a process that causes genes to be expressed from one allele only according to parental origin , the other allele being silent . Diseases can arise when the normally active alleles are not expressed . In this context , low level of expression of the normally silent alleles has been considered as genetic noise although such expression has never been further studied . Prader-Willi Syndrome ( PWS ) is a neurodevelopmental disease involving imprinted genes , including NDN , which are only expressed from the paternally inherited allele , with the maternally inherited allele silent . We present the first in-depth study of the low expression of a normally silent imprinted allele , in pathological context . Using a variety of qualitative and quantitative approaches and comparing wild-type , heterozygous and homozygous mice deleted for Ndn , we show that , in absence of the paternal Ndn allele , the maternal Ndn allele is expressed at an extremely low level with a high degree of non-genetic heterogeneity . The level of this expression is sex-dependent and shows transgenerational epigenetic inheritance . In about 50% of mutant mice , this expression reduces birth lethality and severity of the breathing deficiency , correlated with a reduction in the loss of serotonergic neurons . In wild-type brains , the maternal Ndn allele is never expressed . However , using several mouse models , we reveal a competition between non-imprinted Ndn promoters which results in monoallelic ( paternal or maternal ) Ndn expression , suggesting that Ndn allelic exclusion occurs in the absence of imprinting regulation . Importantly , specific expression of the maternal NDN allele is also detected in post-mortem brain samples of PWS individuals . Our data reveal an unexpected epigenetic flexibility of PWS imprinted genes that could be exploited to reactivate the functional but dormant maternal alleles in PWS . Overall our results reveal high non-genetic heterogeneity between genetically identical individuals that might underlie the variability of the phenotype .
Imprinted genes are functionally mono-allelic in a parent-of-origin specific manner . Genomic imprinting is a non-Mendelian epigenetic form of gene regulation which is germline-inherited since the epigenetic marks are established in the parental gametes without altering the DNA sequence [1] . Compared to most other tissues the brain is enriched in genes showing an imprinted pattern of expression [2] , vulnerable to environmental perturbation [3] and contributing to various neurodevelopmental diseases [4] , [5] . This vulnerability , linked to a plasticity of gene regulation , might also allow a positive adaptation of an organism to a new external environment . It is important to examine situations in which partial loss of imprinting ( LOI ) rescues a mutant phenotype . Understanding the mechanisms underlying this positive effect could lead to therapeutic avenues that manipulate this rheostat function . Necdin ( Ndn ) is an imprinted gene present in both human and mouse , and its maternally inherited allele is normally silenced [6]–[9] . The human NDN gene is located in a large imprinted domain . All the paternally expressed genes from this domain are candidate genes for some of the symptoms of Prader-Willi Syndrome ( PWS ) , an orphan neurodevelopmental genetic disease [10] ( OMIM 176270 ) . The essential clinical diagnostic criteria include neonatal hypotonia and abnormal feeding behavior with a poor suck followed by a hyperphagia , resulting in severe obesity , and behavioral problems [11]–[14] . Breathing deficiency is a significant health concern for many patients and contributes to some cases of sudden death [15] , [16] . Notably , there is considerable variability in symptom severity among patients [11] . Mouse strains with targeted inactivation of single PWS genes have been created , and heterozygous mice with a paternally inherited deficiency ( +m/−p ) are generally considered to be functionally null . Four independent Ndn-deficient mouse lines have been created [9] , [17]–[19] , three of which display PWS associated phenotypes [9] , [17]–[21] including partial early post-natal lethality due to respiratory distress [20] , [22] . In Muscatelli's Ndn-KO mouse model ( named Ndntm1 . 1Mus ) , we observed a high level of phenotypic heterogeneity among the Ndn+m/−p mice within each litter; notably in the incidence and severity of apneas [22] . The stochastic nature of gene expression can generate pronounced phenotypic variations [23] and here we hypothesize that this inter-individual variability , among Ndn+m/−p mice , might result from a “stochastic” activation of the putatively silent maternal allele of Ndn . In this study , we investigate this hypothesis by comparing homozygous mice deleted for both alleles of Ndn ( Ndn−/− ) with heterozygous Ndn ( Ndn+m/−p ) mice . We perform a comprehensive analysis of the in vivo expression and functional role of the Ndn maternal allele . We investigate the genetic context and the mechanism underlying this maternal expression . Finally , we show that the maternal allele of NDN is transcribed and that Necdin protein is present in human post-mortem Prader-Willi brains .
After 36 backcrosses on the C57Bl/6J genetic background , we measured the lethality of Ndn+m/−p mice versus Ndn+/+ mice , both derived from crosses between a wild-type ( WT ) female and a heterozygous male deleted for the Ndn maternal allele ( −m/+p ) . As expected [17] , Ndn+m/−p mice were significantly under-represented at weaning ( 28% reduction , 125 +/+ versus 91 Ndn+m/−p; CHI2 test , P<0 . 01 ) . Furthermore , we confirmed that there was an equivalent number of Ndn+m/−p ( 118 ) versus Ndn+/+ ( 116 ) pups at birth , and 21% lethality between postnatal day ( P ) P0 and P3 in both sexes . However , in a cohort of 75 Ndn−/− mutants , derived from crosses between a Ndn−/− female and a Ndn−/− male , we found a 43% lethality of the Ndn−/− pups ( 32/75 ) , between P1 and P2 , and these pups were visibly cyanotic . Altogether , these data suggest that , due presumably to respiratory deficiency , Ndn−/− newborns are twice as likely to die early compared to Ndn+m/−p newborns . This result is surprising because in theory there is no Ndn expression in either Ndn−/− or Ndn+m/−p pups . Next , we compared breathing pattern between Ndn−/− and Ndn+m/−p mice . Previously , we demonstrated that newborn and young adult Ndn+m/−p mice present an irregular respiratory rhythm with frequent apneas [22] . Importantly , such apneas were more than twice as frequent in Ndn−/− compared to Ndn+m/−p mice ( Table 1 ) . In summary , the respiratory phenotype of Ndn−/− homozygous mice is more severe than in Ndn+m/−p heterozygotes , suggesting a role for the maternally inherited Ndn allele . Since we suspected a role of the maternal Ndn allele in the phenotype of Ndn+m/−p mice , we further investigated the expression of this allele using a specific anti-Necdin antibody in immunoblot analyses of protein extracts from different P1 brains ( Figure S1A ) or from individual E12 . 5 embryos ( Figure S1B ) . We detected a specific signal at the expected size for WT animals but also a fainter signal ( 10–20 fold less intense ) in four of the eight Ndn+m/−p animals , with no signal in Ndn−/− mice . Consistent with our previous study [17] , we did not detect maternal Ndn allele expression by RT-PCR ( Figure S1C ) . We therefore increased the experimental sensitivity using RT-qPCR . We focused on different developmental stages ( E12 . 5 , P1 and adult ) [24] , using total brain tissue as well as brain structures known to highly express Ndn ( hypothalamus ) or to be involved in respiratory function ( pons ) . A total of 258 individual animals on a C57Bl/6J genetic background , including 72 WT , 57 Ndn−/− and 129 Ndn+m/−p , were analyzed . In Ndn−/− mutant mice , no Ndn transcripts were detected irrespective of the brain structures or stages analyzed ( data not shown ) . In WT individuals ( Figure 1A ) , as expected [17] , [24] , we observed higher Ndn expression in P1 brains compared with expression in E12 . 5 embryos . In Ndn+m/−p individuals , maternal Ndn transcripts were detected , but the transcript level was reduced 800 ( P1 brain ) to 1500 ( adult hypothalamus ) -fold compared to WT individuals ( comparing the medians , Figure 1B ) . Interestingly , there was a huge inter-individual variability ( ×100 to ×1000 between the extreme values ) for all Ndn+m/−p mice , irrespective of the stages and tissues tested . We searched for factors that influence the level of transcripts of the Ndn maternal allele . While our results show an absence of a significant effect of the paternal genotype ( Figure 1D , E , F ) , the maternal genotype clearly influences maternal Ndn expression in the Ndn+m/−p offspring . Although in all cases the offspring ( +m/−p ) have inherited a wild-type ( +m ) allele from the mother , offspring from the Ndn+/+ or Ndn+m/−p maternal genotype had a significant three-fold higher level of Ndn maternal expression compared to those from a Ndn−m/+p maternal genotype ( Figure 1D , E , F ) . Importantly , the extensive variability of Ndn maternal expression is also positively correlated with both those maternal genotypes . In contrast , litters issued from Ndn−m/+p mothers showed both a lower level of Ndn expression and an absence of variability in the Ndn+m/−p offspring ( Figure 1D , E , F ) . In addition , there is a gender-specific effect on the Ndn maternal expression in Ndn+m/−p offspring , with females expressing two-fold more Ndn expression compared to males ( Figure 1G ) . Finally we compared this maternal expression in Ndn+m/−p offspring from a C57Bl/6J or a 129Sv/Pas genetic background . In both mouse strains , a similar level of Ndn maternal expression was observed ( Figure 1H ) . We conclude that an extremely low but specific transcription of the maternally inherited Ndn allele in Ndn+m/−p individuals is found in at least two mouse strains ( C57Bl/6J , 129Sv/Pas ) . Transcript numbers are highly variable irrespective of the developmental stage or the brain structure analyzed , even among littermates . Finally , the quantity of maternal Ndn transcripts depends significantly on the maternal genotype and on the gender . We asked whether maternal Ndn expression was due to: 1 ) low but homogeneous expression in all tissues and/or 2 ) reduced but focal expression in specific structures and cell types . At E12 . 5 , using immunohistochemistry ( IHC ) and in situ hybridization ( ISH ) on frozen serial sections , Necdin protein and transcripts were detected in the same structures of four out of nine Ndn+m/−p embryos ( Figure 2 ) . Importantly , no protein or transcripts were detected in Ndn−/− individuals ( Figure S2 ) . Expression of the maternally inherited Ndn allele was detected in a restricted number of cells of specific nervous structures in which the paternally inherited Ndn allele is normally expressed in WT animals ( Figure 2 A , B , C , D , E ) . Interestingly , the cerebral cortex , the tongue and the myotome , which show expression of the paternal allele in WT , do not express the maternal Ndn allele ( data not shown ) . In contrast to WT E10 . 5 embryos , no maternal Ndn expression was detected in Ndn+m/−p E10 . 5 embryos ( n = 9 ) ( Figure S3 ) . At P1 , a stage when expression normally peaks [24] , we detected Necdin protein by IHC in the brain of Ndn+m/−p newborns ( n = 6 ) ( Figure 3 ) . Necdin presented a similar expression pattern in Ndn+m/−p adults ( Figure S4 ) . At both developmental stages , this expression was restricted to a limited number of cells in several , but not all nuclei that express Ndn in WT animals , such as the hypothalamic ( Figure 3B ) and the raphe nuclei ( Figure 3C , D ) . Thus , at the anatomical level , using ISH and IHC , we conclude that the Ndn maternal allele is expressed in a subset of Ndn+m/−p individuals and , compared to WT mice , is restricted to a limited population of cells in specific nervous system structures . Noticeably , there is considerable inter-individual variability , even between littermates ( Figure 1 and data not shown ) . We addressed the question of intra-individual variation by studying the raphe nuclei , a structure defined by 5HT-expressing neurons , all of which express Necdin in WT mice [22] . We double immunostained P1 brains using anti-Necdin and anti-5HT antibodies and determined the number of 5HT/Necdin positive neurons in the different raphe nuclei ( B1 to B9 ) of WT , Ndn+m/−p and Ndn−/− newborns ( Figure 4A ) . We confirmed both inter-individual and intra-individual variation in the number of 5HT/Necdin double positive neurons in Ndn+m/−p raphe nuclei . For instance , in the same individual , 78% of 5HT positive neurons in B1/B2 raphe nuclei were Necdin positive although in other raphe nuclei no Necdin expression was detected ( Figure 4A ) . We conclude that there is also intra-individual variation in the expression of maternal Ndn allele in the raphe nuclei . Previously , we observed alterations in the 5HT system [22] in Ndn+m/−p mice . Here , we analyzed the cellular defects in Ndn−/− newborn mice ( n = 8 ) in comparison with Ndn+m/−p ( n = 18 ) and WT newborns ( n = 9 ) . Using 5HT immunolabelling , we counted the number of 5HT neurons in the B1/B2 raphe nuclei ( Fig . 4B ) . We found a significant 28% reduction ( WMW test , P<0 . 001 ) in the number of 5HT-expressing neurons between Ndn−/− ( 1306 ( 1204 , 1337 ) ; n = 8 ) and WT newborns ( 1807 ( 1738 , 1882 ) ; n = 9 ) . Interestingly , in the B1/B2 raphe nuclei compared to WT mice , the Ndn+m/−p individuals that expressed Necdin , with a mean of 46% of 5HT neurons Necdin-positive ( Ndn+m/−p ( Ndn+ ) , Figure 4B ) , had only a 8% reduction ( WMW test , P<0 . 001 ) in the number of 5HT-expressing neurons ( 1666 ( 1489 , 1733 ) ; n = 9 ) . In contrast , the Ndn+m/−p individuals ( Ndn+m/−p ( Ndn− ) , Figure 4B ) that do not show Necdin expression had a significant 28% reduction ( WMW test , P<0 . 001 ) in the number of 5HT-expressing neurons ( 1313 ( 1258 , 1335 ) ; n = 9 ) similar to the results observed in Ndn−/− P0 mice . Thus expression of the maternal Ndn allele in Ndn+m/−p individuals correlates with an increased number of 5HT-expressing neurons . We next asked whether the low level of maternal Ndn expression was also present in WT mice . In order to discriminate between paternal and maternal allele-specific Ndn expression in WT mice , we identified mouse strains carrying transcribed polymorphisms in the Ndn gene . Three such polymorphisms ( two SNPs in the 3′-untranslated region ( UTR ) and one 5bp indel in the 5′-UTR ) were identified between Mus musculus ( C57BL/6J ) and Mus spretus strains . First , to analyze the SNPs , we performed two quantifications of allele-specific expression by pyrosequencing ( QUASEP ) assays on RT-PCR products from F1 brains of six pups with a C57BL/6J mother and Mus spretus father , and did not detect expression of the maternal ( C57BL/6J ) Ndn allele in these brain samples ( data not shown ) . To further increase the sensitivity for detection of maternal Ndn transcripts , we designed specific TaqMan probes distinguishing between the presence and absence of the 5 bp indel in the 5′-UTR and used RT-qPCR for allele-specific quantification . However , this assay also did not reveal any Ndn transcripts from the C57BL/6J maternal allele in ( C57BL/6J×Mus spretus ) F1 brains from 32 pups ( Figure S5 ) . We conclude that in this wild-type mixed genetic context , we do not detect any expression of the maternal C57Bl/6J Ndn allele . The Ndn+m/−p heterozygous mice described by Gerard et al [9] ( named Ndntm2Stw ) present a more severe phenotype with in particular a higher lethality at birth compared to Ndntm1 . 1Mus+m/−p mice . We failed to detect expression of the Ndn maternal allele in Ndntm2Stw+m/−p embryos ( n = 8 ) at two developmental stages ( E12 . 5 and E14 . 5 ) , using the IHC and ISH approaches ( Figure S6 and data not shown ) . Importantly , in Ndntm2Stw mice , the Ndn coding sequence has been replaced by the β-Galactosidase sequence , and the Ndn promoter and regulatory sequences have been retained allowing β . Gal expression from the paternal allele [9] . In contrast , in Ndntm1 . 1Mus mice , the promoter and the first two thirds of the Ndn coding sequence were replaced with a loxP site . The complete lack of Ndn maternal expression in the Ndntm2Stw+m/−p mice could suggest that presence of the active paternal Ndn promoter suppresses expression of the Ndn maternal allele . We propose that the Ndn maternal allele is expressed only when the paternal promoter is absent or silenced , consistent with the absence of expression of the maternally inherited Ndn allele in WT mice ( Figure S10A ) . To further explore this question , we created a transgenic mouse line ( TG45 named TG ) containing a modified Bacterial Artificial Chromosome ( BAC ) in which the Ndn coding sequence was replaced by the eGFP sequence under the control of the Ndn promotert; this BAC transgene is present in one or two copies and its expression is not regulated by imprinting mechanism ( Figure S7 and data not shown ) . In these mice , the expression of eGFP was restricted to the brain regions in which Ndn is normally expressed ( Figure 5 and data not shown ) . We then studied this eGFP expression in newborn WT , Ndn+m/−p , Ndn−/− hypothalamus ( using the Ndntm1 . 1Mus strain ) and in the hypothalamus from a mouse line ( Ndn++ ) in which Ndn is over-expressed ( Figure 5 and Figure S10B ) . We performed colabelling and observed an inverse correlation between the intensity of Necdin immunolabeling and eGFP fluorescence on coronal brain sections ( Figure 5 ) . In the absence of Ndn expression ( Ndn−/− , TG+ ) the number of eGFP-positive cells was the highest compared to eGFP-positive cells when Ndn is over-expressed ( Ndn++ , TG+ ) ( Figure 5 ) . This result was quantified by immunoblotting ( Figure S8A ) . Finally , in ( WT , TG+ ) mice , we independently quantified for each cell the eGFP and Necdin signals in two hypothalamic nuclei ( Figure S8 B , C ) . For both structures , we observed two distinct populations of cells and conclude that approximately half of the cells expressed Necdin while the other half expressed eGFP; only very few cells co-expressed eGFP and Necdin ( Figure S8D ) . Thus , at the cellular level , the eGFP expression level is inversely correlated with Necdin expression level . Our results are consistent with a model whereby two Ndn alleles in the same cell , both of which include at least the Ndn promoter and regulatory sequences and neither of which is silenced by imprinting , triggers allelic exclusion at the transcriptional level favouring the expression of one allele only per cell ( Figure S10 ) . Variation in DNA methylation at the DMRs of imprinted genes has been reported in different tissues , importantly in brain , and might be a source of gene expression and phenotypic variations [25] . We therefore studied DNA methylation in a secondary DMR ( 42 CpGs ) , previously shown to be correlated with imprinted regulation of Ndn expression [7] , [26] ( Figure S9 ) . We found no major changes in methylation on the Ndn maternal allele in Ndn+m/−p brains . However , methylation of this DMR occurs after the blastula stage and our failure to detect modifications to methylation in this DMR could be because only a few neurons express the maternal Ndn allele , and this escapes our global brain analysis . We assessed whether the expression of the maternal Ndn allele observed in heterozygous Ndn+m/−p mice also occurs in PWS patients . Using a specific human NDN RNA probe and an anti-Necdin antibody , we performed an ISH and IHC on hypothalamic sections obtained from brains from two adult PWS patients ( one with a deletion and one with a maternal disomy ) and one PWS infant ( 9 months old with a deletion ) ; age and sex matched control individuals were included as positive controls ( Table S1 ) . In all patients , we found NDN transcripts and protein in the paraventricular and supra optic nuclei ( Figure 6 ) . We confirmed NDN mRNA expression in five more adult PWS patients ( 25–64 years of age ) and one PWS child ( 6 months old ) . Expression of NDN also occurred in the cortex of PWS patients ( data not shown ) . The results contradict the widely accepted assumption that in PWS patients the maternal allele is totally silenced in the brain . These findings are in full agreement with the results obtained in our heterozygous Ndn+m/−p mice .
Prior to this study , it was widely accepted that only the paternal alleles of PWS candidate genes are expressed , the maternal alleles being totally silenced . However , in brains of mice , with a deletion of the imprinting center , an incomplete silencing of paternally inherited PWS genes as well as a low level of expression of maternal alleles of PWS genes , was reported but not investigated [27] . LOI was also observed in lymphoblasts of two PWS patients with a deletion and two atypical PWS patients with a maternal disomy [28] , [29] , but these studies were not extended to include expression profiles in the brain . LOI has been described in other contexts , particularly in some cancers [30] . Our study addresses for the first time the robustness of silencing of the maternal alleles of PWS candidate genes in brain . Our results show that in both mice and humans , in the absence of the paternally inherited Ndn gene , the maternal Ndn allele is expressed in the brain at very low level but sufficiently to allow Necdin protein production . A similar mechanism might be hypothesized for any of the PWS genes in PWS patients . For example , the imprinted Magel2/MAGEL2 PWS gene , showed a similar loss of imprinting in Magel2+m/−p heterozygous mice and PWS human brains ( F . M . and D . S . unpublished data ) . Significantly , the high variability of expression of maternal alleles of these genes might explain the large degree of heterogeneity in the severity of PWS symptoms . The two-fold reduction of post-natal mortality in Ndn+m/−p mice compared to Ndn−/− mice , suggests that even the low level of maternal Necdin protein is sufficient to rescue 50% of the mice , in comparison with the Ndn−/− mice . Nevertheless 50% of Ndn−/− individuals survive suggesting that another compensatory system is activated when the level of Ndn expression is null or very low in Ndn+m/−p mice . A surprising degree of inter-individual variability was observed in the number of Ndn transcripts amongst Ndn+m/−p mice . The degree of maternal Ndn expression is correlated with the severity of the phenotype , in that the number of apneas is significantly increased in Ndn−/− mice compared to Ndn+m/−p mice . Previously , we published that those apneas might be correlated with an alteration of the 5HT system [22] . Here we showed that the number of 5HT-expressing neurons is reduced by 28% in Ndn−/− compared to WT mice while Ndn+m/−p mice are divided in two distinct populations with 30% and 10% fewer 5HT-neurons respectively . The lowest reduction ( 10% ) of 5HT-expressing neurons is observed in those Ndn+m/−p individuals co-expressing Necdin in 5HT neurons . These data support a link between the expression of the Ndn maternal allele and the degree of survival , the severity of apneas and the number of 5HT neurons in the B1/B2 raphe nuclei . Ndn maternal expression presents a high inter-individual variability ( 1 to 3 orders of magnitude ) , even among Ndn+m/−p individuals from the same litter , irrespective of the age and brain structure analyzed . Intra-individual variability of Ndn expression was also detected in the brain structures . This expression is limited to some , but not all , of the brain regions that normally express Ndn with no evidence of ectopic expression . In those brain regions the number of neurons expressing Ndn is clearly less than in wild-type animals and variability of Ndn expression amongst the Ndn+m/−p offspring was linked to both maternal genotype and gender , being additive factors . A Ndn+m/−p mouse with a +/+ or Ndn+m/−p mother ( a mouse who has inherited a wild-type Ndn allele from her grandmother ) is predisposed to the highest level of expression and to a greater inter-individual variability , a phenomenon referred to as transgenerational epigenetic inheritance [31] . In contrast , paternal genotype has no impact . Furthermore maternal Ndn allele expression was two-fold higher and more variable in female mice compared to male mice . This may reflect the increased genetic variability in females: some genes escaping X inactivation , such as Jarid1c , which codes for a histone demethylase [32] , showing higher expression in females . This could explain our observations concerning maternal allele Ndn expression . Alternatively or additionally , female-specific hormones could be involved . An interesting observation resulting from transcriptome profiling is the very high variability between individuals in steady state levels of a range of mRNAs , often reaching an order of magnitude [33] . This might explain why the penetrance of a given genotype is often incomplete [23] , [31] , [33] . This type of epigenetic phenomenon might also be involved in the variable expression of the maternal Ndn gene and consequently might lead to survival of some Ndn+m/−p mice . Nevertheless , even in Ndn−/− mice the penetrance of the phenotype ( postnatal lethality and apneas ) is not complete , suggesting that another mechanism involving a “compensatory pathway” takes place . This compensatory pathway might result from an increase of a gene expression linked to the lack of Ndn expression or might also result from the stochastic variability in gene expression described above [33] that occurs independent of the state of Ndn expression . The lack of detection of expression of the C57Bl/6J maternal allele in WT mice ( with a paternal M . Spretus allele ) suggests that expression of the Ndn maternal allele is associated with the absence of an active paternal Ndn promoter , as confirmed by our study of the Ndn+m/−p Ndntm2Stw embryos that did not express the maternal allele of Ndn . Similarly , in a third Ndn-KO mouse model , in which the Ndn-promoter also drives β-gal expression [18] , no maternal Ndn transcripts were detected by RT-qPCR [34] in Ndn+m/−p mice . Previously , Chamberlain et al . observed a low level of Ndn maternal expression only in the absence of an active paternal PWS-imprinting center , and suggested a trans effect where the paternal PWS-IC acts on the maternal allele [27] . Furthermore , our data suggest that , even in the absence of imprinted regulation of Ndn , as is the case for the Ndn-eGFP BAC transgene , it appears that there is a transcriptional regulation predisposing to a monoallelic expression of Ndn . This result might be explained by promoter competition for transcriptional activators , or a mechanism involving physical contact in trans between promoters [27] , [35]–[37] . Given the extremely low level of Ndn transcripts in +m/−p mice as estimated by RT-qPCR it is surprising that the protein was detectable by immunohistochemistry and by Western blot . Importantly , the absence of antibody staining on samples from −/− mice ruled out the possibility of cross-reactivity with proteins sharing epitopes with Necdin . Until relatively recently , it has been assumed that transcript abundance is the main , although not the only , determinant of protein abundance . Experiments aimed at addressing this question have lead to an emerging body of evidence changing this view and , in every organism that has been examined to date at a global level , steady-state transcript abundance only partially predicted protein abundance [38] . This lack of correlation suggests a strong regulatory role for all processes downstream of transcription . Furthermore , it has been shown that in many situations , transcription , translation and degradation are often extensively coupled and regulate each other through feedback loops . This coupling might enhance responsiveness to the environment and might help reduce inter-cellular variability in gene expression , which is by nature a stochastic event [39] . Collectively , these results suggest that very low level expression of PWS maternally silenced genes might be sufficient to alleviate specific PWS symptoms [23] . Importantly , we show that the quantity of Ndn transcripts is not , at least in neurons , a good indicator of its protein level and hence its functional importance [38] . An understanding of the context in which the Ndn maternal allele might be transcribed is an important step towards the development of a pharmacological therapy to trigger and/or increase the expression of this maternal allele in PWS patients . Furthermore , our results provide a further indication of the high non-genetic heterogeneity between genetically identical individuals that might , in this case , underlie LOI and contribute to variability in the phenotype [40] .
Mice were handled and cared in accordance with the Guide for the Care and Use of Laboratory Animals ( N . R . C . , 1996 ) and the European Communities Council Directive of September 22th 2010 ( 2010/63/EU , 74 ) . Experimental protocols were approved by the institutional Ethical Committee guidelines for animal research with the accreditation no . B13-055-19 from the French Ministry of Agriculture . Ndn deficient mice were maintained on the C57BL/6J background and the paternal mutation was transmitted by crossing Ndn−m/+p males with C57BL/6J WT females ( from Janvier Company ) . In parallel , since the Ndn-KO allele was created using a 129/SvPas ES cell line , we maintained the mutation via a maternal transmission on the 129/SvPas genetic background using Charles River male mice . All Ndn mice were genotyped by PCR as previously described [21] . Genotype of Ndn−/− mice was confirmed by a secondary intra-deletional PCR whose primers were: 5′-GATCCGAAGGCGCAGACATG-3′ and 5′-CTGCCCATGACCTCTTTCAC-3′ generating a 420 bp fragment indicating the presence of Ndn WT allele . The Ndn ++ over-expressing mouse line is the Magel2 KO ( +m/−p and −/− ) mouse line created previously in our team . Consequently , it is an over-expression of the endogenous Ndn gene rather than being a transgenic mouse . Magel2 being imprinted , closed to Ndn and belonging to the MAGE family gene , as Ndn . We observed this overexpression at the transcript and protein level and we estimated , by western blot quantification , the level of overexpression ( a factor of 1 . 7 fold ) . They were also maintained onto C57BL/6J background . Importantly , all the mouse lines used in this study , excepted the Ndntm2Stw+m/−p mouse line , have been created in our laboratory and maintained on pure genetic background . The Ndntm2Stw+m/−p mouse have also been bred onto C57Bl/6 for over 30 generations in Wevrick's laboratory . ” P1 brain and E12 whole embryos were rapidly dissected and crushed in lysis buffer as previously [41] . For each sample , proteins ( 30 mg ) were separated on a 12% SDS-PAGE and transferred onto nitrocellulose membranes ( Protran Whatman , Dutscher ) . Membranes were incubated overnight with a rabbit polyclonal antibody against Necdin ( Upstate; 1∶1000 ) or with a rabbit polyclonal antibody against GFP ( Sigma , G1544 ) and subsequently with an anti-rabbit horseradish peroxydase antibody ( GE Healthcare , Buckinghamshire , UK; 1∶3000 ) . In both experiments , membranes were reprobed using a mouse anti-α-Tubulin antibody ( Sigma , T6074 ) . For Necdin immunolabeling was visualized by enhanced chemiluminescence . For GFP immunolabeling was visualized by Gbox ( Syngen ) . Quantification was performed using ImageJ . We performed plethysmography in weight-matched littermate mice that were 6 weeks old , unrestrained and unanesthetized . Spontaneous breathing activities were recorded in normoxic conditions using whole-body plethysmograph ( EMKA Technologies , Paris , France ) . After a 30 min period of stabilization in the apparatus , respiratory parameters were calculated breath-by-breath during a 30 min period of measurement . The mean of each parameter was automatically calculated from this 30 min period of measurement using EMKA technologies Datanalyst software . Apneas have been defined here as an absence of a respiratory signal during at least three respiratory cycles in resting conditions . Classical RT-PCR was performed as previously described [17] . For RT-qPCR , mice were sacrified at E12 . 5 , P1 or as adults . Whole embryos , whole P1 brains , P1 pons and P1 or adult hypothalamus tissues were rapidly collected and frozen in liquid nitrogen prior to RNA isolation using standard conditions . Subsequently , total RNA samples were incubated with DNase ( TURBO DNA-free; Ambion ) . Messenger RNAs from 1 µg of total RNAs were reverse-transcribed in a total volume of 20 µL using the M-MLV , reverse transcriptase RNAse H minus , point mutant ( Promega ) and oligod ( T ) 15 in the presence of a synthetic external , heterologous and noncompetitive poly ( A ) Standard RNA ( SmRNA ) used to calibrate the reverse transcription [42] ( patent WO2004 . 092414 ) . At the end of the RT , total volume was brought up to 100 µL and real-time PCR was performed using the Rotorgene System ( Qiagen ) to determine the number of SmRNA and Ndn cDNA molecules in 5 µL of the RT product . The specific forward and reverse primers were designed using “Universal Probe Library” software ( Roche Diagnostics ) in the region deleted in the Ndn KO-allele . The sequences of the primer pair used were: Necdin-Forward 5′-AACAACCGTATGCCCATGA-3′ , Necdin-Reverse 5′-CTTCACATAGATGAGGCTCAGGAT-3′ ( 60 bp ) . The primer sequences and the quantification conditions of calibrator cDNAs ( Standard cDNAs ) are protected by the patent WO2004 . 092414 . To discriminate specific from nonspecific cDNA products , a melting curve was obtained at the end of each run , by a slow temperature elevation up to 98°C ( 0 . 1°C . s-1 ) . Before RT , absence of traces of genomic DNA in the purified total RNA samples was ruled out by real-time PCR of the non-deleted Ndn sequence . Quantification cycles were converted into the number of cDNA copies using the quantification curve specific for each primer pair that had been previously established from serial dilutions of purified PCR products . The equation of the calibration curve for NDN cDNA was performed in four replicates for each dilution ranging from 10 to 1×109 copies : Ct = −3 . 3417 Log [cDNA]i+39 . 049 , r2 = 0 . 9988 . No amplification was obtained in Ndn−/− individuals only . In the other mice , the lowest and highest copy numbers quantified were 92 and 4 , 638 , 062 , respectively . For each sample , the number of Ndn cDNA copies was normalized according to relative efficiency of RT determined by the standard cDNA quantification . Finally , gene expression was expressed as the cDNA copy number quantified in 5 µL aliquot of RT product . Specificity of Necdin protein detection was controlled on tissues from Ndn−/− animals . Fixed brains was dissected , cryopreserved and sectioned ( 14 µm ) using a cryostat ( Leica CM3050S ) . Embryos and post-natal mice ( P1 ) were sacrificed and treated as previously [21] . Antibodies used were: rabbit polyclonal anti-Necdin ( 07-565; Millipore , Bedford , MA , USA; 1∶500 ) , mouse monoclonal anti-GFP ( Interchim , NB600-597; 1∶500 ) , goat polyclonal anti-5HT ( Immunostar , 20079; 1∶300 ) . Sections were washed twice in PBS and incubated with Hoechst ( 33258 , Sigma; 1∶2000 ) and corresponding fluorochrome-conjugated secondary antibodies , goat anti-rabbit Alexa Fluor 488 or Alexa Fluor 555 ( Molecular Probes , Invitrogen; 1/500 ) , goat anti-mouse Alexa Fluor 488 ( Molecular Probes , Invitrogen; 1/500 ) , donkey anti-goat Cy3 ( Chemicon , AP180C; 1/1000 ) diluted in the blocking buffer without BSA . Sections were examined on a Zeiss Axioplan 2 microscope with an Apotome module . Quantification of labeled cells was performed using ImageJ . For quantification of immunofluorescence , images were acquired using a confocal microscope ( SP5-X , Leica ) , z stacks of 70 µm were performed for each image , and analyzed using ImageJ . All Ndn in situ hybridization experiments for the study of Ndn gene expression were performed on serial slices of those used for immunohistochemistry and performed as previously [43] . Specificity of Ndn mRNA detection was controlled on tissues from Ndn−/− animals and with the sense control riboprobes . A peroxidase-conjugated anti-digoxigenin-POD ( 1∶1250 ) antibody ( Roche ) was used to detect the Ndn hybridized riboprobe , visualized using a tyramide signal amplification ( TSA-plus Biotin Kit , Perkin Elmer ) . We could identify two transcribed Single Nucleotide Polymorphisms ( tSNPs ) in the 3′-UTR of Ndn to discriminate between the C57BL/6J and Mus spretus alleles . To determine allele-specific transcription levels , we performed QUASEP and RT-qPCR with allele-specific TaqMan probes on cDNA of C57BL/6J×Mus spretus F1 brains from P10–P14 pups . All RNA samples were treated with DNaseI ( Agilent ) to minimize any risk of contamination with genomic DNA . Subsequently , 2 µg of high-quality total RNA were reverse transcribed into cDNA ( SuperScript III First Strand Synthesis System , Invitrogen ) and oligo ( dT ) -priming according to manufacturer's instructions . The QUASEP assays were designed using the PyroMark Assay Design Software 2 . 0 ( Qiagen ) . PCR was performed with the FastStart High Fidelity PCR System ( Roche ) according to manufacturer's recommendations using the cDNA of C57BL/6J×Mus spretus F1 brains from 6 P10–P14 mice . Pyrosequencing was done on a PSQ 96MA Pyrosequencing System ( Qiagen ) with a sequencing primer ( Table S2 ) and PyroGold SQA reagents ( Qiagen ) . Data were analyzed with the PSQ 96MA 2 . 1 . 1 software ( Qiagen ) as previously described [44] . For allele-specific RT-qPCR , we used the 5 bp indel in the 5′-UTR of Ndn to design TaqMan probes specific for C57BL/6J and Mus spretus , respectively [45] , [46] . Quantitative PCR was performed on an ABI 7500 Fast Real time PCR System using the cDNA of C57BL/6J×Mus spretus F1 brains from 32 P10–P14 mice . Briefly , the 20 µl reaction contained 10 µl TaqMan Fast Universal PCR Master Mix ( 2× ) , 3 . 6 µl 5 µM combined forward ( C57BL/6J: 5′-CTTCCTCTGCTGGTCTCCAC-3′ , Mus spretus: 5′-CTTCCTCTGCTGGTCTCCAC-3′ ) and reverse ( C57BL/6J: 5′-GGGTCGCTCAGGTCCTTACT-3′ , Mus spretus: 5′- GGGTCGCTCAGGTCCTTACT-3′ ) primers ( 0 . 9 µM ) ; 2 µl of each 2 µM TaqMan probe ( C57BL/6J: FAM-CTCCAAGCCGCATCGGTCCTGCTC-BHQ1 , Mus spretus: ATTO550-CTCCAAGCCGCATCGCATCGGTCC-BHQ2; 0 . 2 µM ) and 2 . 4 µl cDNA . The qPCR thermal profile consisted of 95°C for 10 min , followed by 48 cycles of 95°C for 30 s and 60°C for 30 s . Real-time PCR data were analyzed with ABI SDS 2 . 0 . 6 software . Unfertilized oocytes and blastocysts were collected from C57BL/6 superovulated females and directly embedded in agarose beads for bisulphite treatment as previously described [47] . Sperm was recovered from the epididymis . Adult brain and kidney were dissected from interspecific M . spretus X M . musculus F1 mice or from Ndn+m/−p mice and DNA extracted according to standard techniques . Bisulphite treatment of HindIII-digested adult brain , kidney and sperm genomic DNAs was carried out as described [48] . Oocyte and blastocyst DNAs were treated as described [47] . A semi-nested PCR was used to amplify regions A ( CpG sites 1 to 20 ) and B ( CpG sites 21 to 42 ) from bisulphite treated DNA samples . Primers used to amplify region A were: 5′-TGTGTTATATAGGAGATTAGG-3′ ( outside forward; first and second rounds ) , 5′-AAACTACCATAAAACCTT-3′ ( outside reverse ) and 5′-CTATCCTACATCTCACAA-3′ ( inside reverse ) . Primers used to amplify region B were: 5′-ATTGTGAGATGTAGGATAG-3′ ( outside forward; first and second rounds ) , 5′-CCATAACCTCTTTCACCATA-3′ ( outside reverse ) , and 5′-AAACTACCATAAAACCTTC-3′ ( inside reverse ) . PCRs were performed in 50 µl reactions containing 1 . 25% DMSO , 25 pmole of each primer , 0 . 2 mM dNTPs , 2 . 5 mM MgCl2 , 1× PCR Buffer and 2 . 5 U Q-BioTaq DNA Polymerase ( Quantum Appligene , Germany ) . Second round PCRs were performed using 1 µl of the purified primary PCR products . PCR cycles were 5 min at 94°C followed by 10 cycles of 30 s at 94°C , 30 s at 58°C , 25 s at 72°C , and by 25 cycles of 30 s at 94°C , 30 s at 58°C , 25 s plus 5 s at each cycle at 72°C , and 7 min at 72°C . Purified PCRs products were cloned into the pGEM-T Easy TA Vector ( Promega ) and sequenced using standard methods . For oocytes and blastocysts , PCRs were performed on samples prepared from at least three different batches of oocytes and blastocysts . Identical clones derived from oocytes and blastocysts secondary PCRs were considered as derived from one single allele and represented only once . The BAC603M20 ( Research Genetics; referred as BAC109 [43] ) contains a 104 kb NotI insert including the Ndn gene . BAC109 was modified by homologous recombination in E . coli as described [49] in order to replace the Ndn open reading frame ( ORF ) by the eGFP ORF . Cesium chloride gradient purified BAC DNA was microinjected in the pronucleus of C57BL/6× CBA mouse zygotes . Founders containing the BAC transgene were identified by amplifying an eGFP fragment by PCR . Transgenic founders were maintained on C57BL6 genetic background . Transgene copy numbers were determined by Southern blot of BglII digested genomic DNA hybridized with PCR probes . Hypothalamic material from 3 PWS patients and from controls , matched for age , sex , postmortem delay and fixation time were obtained through The Netherlands Brain Bank ( NBB , Director Dr . I . Huitinga ) . Clinicopathological details are given in Supplementary Table S1 . Sections throughout the hypothalamus were collected at 1200 µm intervals and mounted and pretreated as previously [50] , with 2 µg/ml of proteinase K . For the detection of NDN mRNA we hybridized the sections with a 2000 ng/ml DIG-labeled RNA probe , complementary to bp1258–1578 of the human NDN mRNA ( NM_002487 . 2 ) . Hybridization and stringency washes were performed as previously described [51] at 60°C . Anti-DIG-Alkaline phosphatase-fab fragments ( Roche ) , diluted 1∶3000 in buffer 1 ( 100 mM Tris , 150 mM NaCl pH 7 . 5 ) were used to detect DIG labeled RNA hybrids [50] . Specificity of the hybridization signal was verified by comparison with sections processed with sense probe under identical conditions . For IHC , 6 µm sections of hypothalamus tissue , containing SON , PVN , and INF were collected , mounted and microwaved in Citrate Buffer as previously [50] . Necdin was detected with rabbit IgG , anti-Necdin ( 07-565; Millipore , Bedford , MA , USA ) diluted 1∶500 in Supermix ( SUMI: 0 . 25% gelatin ( Merck ) ( w/v ) , 0 . 5% Triton X-100 in TBS , pH 7 . 6 ) for 1 hour at RT , followed by an overnight incubation at 4°C . Detection of Necdin immunoreactivity was performed according to the ABC method described before [52] . Antibody specificity was confirmed by the absence of ICC staining in the human hypothalamus after omission of the first antibody from the staining protocol . Nonparametric statistical tools ( Sigmastat software ) or exact statistical tools ( StatXact software ) were used depending on the size of the sample ( n ) . All tests are two-tailed tests . In the results , values are indicated as following: Mean±SD or ( Q2 ( Q1 , Q3 ) , n , P value ) where Q2 is the median , Q1 is the first quartile and Q3 is the third quartile . Mann Whitney t-test or Wilcoxon-Mann-Whitney test ( WMW in the text ) are used . The level of significance was set at a P-value less than 0 . 05 . | Genomic imprinting is a process that causes genes to be expressed from only one of the two chromosomes , according to parental origin , the other copy of genes being silent . Prader-Willi Syndrome ( PWS ) is a neurodevelopmental disease involving imprinted genes , including NDN , which are only expressed from the paternally inherited chromosome , the maternally inherited copy of the gene normally being silent . Here we show that , in absence of the paternal Ndn copy only , the maternal Ndn allele is expressed at an extremely low level with a high degree of heterogeneity . The level of this expression is dependent on both the sex of the offspring and the genotype of the mother . In about 50% of mutant mice , this expression reduces birth mortality and severity of the breathing deficiency , showing a functional role of this low expression . Importantly , specific expression of the maternal NDN allele is also detected in post-mortem brain samples of PWS individuals . Our data reveal an unexpected epigenetic flexibility of PWS imprinted genes that could be exploited to reactivate the functional but dormant maternal alleles in PWS . Overall our results reveal high non-genetic heterogeneity between genetically identical individuals that might contribute to variability in the phenotype . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Stochastic Loss of Silencing of the Imprinted Ndn/NDN Allele, in a Mouse Model and Humans with Prader-Willi Syndrome, Has Functional Consequences |
Ebola virus ( EBOV ) is a highly pathogenic filovirus that causes hemorrhagic fever in humans and animals . Currently , how EBOV fuses its envelope membrane within an endosomal membrane to cause infection is poorly understood . We successfully measure cell-cell fusion mediated by the EBOV fusion protein , GP , assayed by the transfer of both cytoplasmic and membrane dyes . A small molecule fusion inhibitor , a neutralizing antibody , as well as mutations in EBOV GP known to reduce viral infection , all greatly reduce fusion . By monitoring redistribution of small aqueous dyes between cells and by electrical capacitance measurements , we discovered that EBOV GP-mediated fusion pores do not readily enlarge—a marked difference from the behavior of other viral fusion proteins . EBOV GP must be cleaved by late endosome-resident cathepsins B or L in order to become fusion-competent . Cleavage of cell surface-expressed GP appears to occur in endosomes , as evidenced by the fusion block imposed by cathepsin inhibitors , agents that raise endosomal pH , or an inhibitor of anterograde trafficking . Treating effector cells with a recombinant soluble cathepsin B or thermolysin , which cleaves GP into an active form , increases the extent of fusion , suggesting that a fraction of surface-expressed GP is not cleaved . Whereas the rate of fusion is increased by a brief exposure to acidic pH , fusion does occur at neutral pH . Importantly , the extent of fusion is independent of external pH in experiments in which cathepsin activity is blocked and EBOV GP is cleaved by thermolysin . These results imply that low pH promotes fusion through the well-known pH-dependent activity of cathepsins; fusion induced by cleaved EBOV GP is a process that is fundamentally independent of pH . The cell-cell fusion system has revealed some previously unappreciated features of EBOV entry , which could not be readily elucidated in the context of endosomal entry .
Ebola virus ( EBOV ) outbreaks continually occur and up to 90% of those infected die; currently there are no approved vaccines or antiviral therapeutics against the virus [1 , 2] . EBOV initiates infection by fusion from within endosomes . Experimentally , endosomal interiors are difficult to control , but systems that track the entry of several other viruses into cells have been developed and employed [3 , 4 , 5 , 6] . Historically , these methods have relied on fusion of infectious virus or pseudovirus within cells; cell-cell fusion has not been among the systems in use for EBOV . It is surprising that a cell-cell fusion system has not been developed , as the processing of the Ebola fusion protein , GP , and other conditions necessary for fusion have been elaborated [7] . ( Some years ago there was an isolated report of EBOV GP-mediated cell-cell fusion , but this study has not been followed up by any other laboratory , including the original [8] ) . Cell-cell fusion has several important advantages over intracellular fusion assays , including complete control of the aqueous solution bathing the ectodomain of the fusion protein . In the present study we describe a direct and sensitive system to measure EBOV GP-mediated cell-cell fusion with high time resolution , thereby providing fusion kinetics . The system exhibits the well-known central properties of EBOV entry , providing strong support for the utility of the cell-cell fusion system to explore mechanisms of EBOV entry that are not possible or practical with whole infectious virus . EBOV GP is a prototypic class I viral fusion protein [9] . It is synthesized as a homotrimer; each monomer is cleaved into GP1-GP2 subunits by proteases within the Golgi apparatus [10 , 11] . The GP1 subunit is responsible for binding to the intracellular receptor Niemann Pick type C1 , ( NPC1 ) and possibly to other molecules [12] , and the GP2 subunit is responsible for membrane fusion [13 , 14 , 15 , 16 , 17 , 18 , 19] . The two subunits of each monomer remain linked through a disulfide bond and a multitude of weak interactions [9 , 20 , 21 , 22] . After endocytosis of the virus , the GP1 subunit is cleaved by the endosomal proteases cathepsin B and/or L [7 , 23 , 24 , 25 , 26] , while remaining attached to GP2 [9] , and then binds to NPC1 [14 , 15] . The low pH within endosomes is necessary for viral fusion . But it has not been known whether low pH directly triggers fusion by causing conformational changes in GP or whether it augments fusion by increasing the activities of the cathepsins [7 , 25] . After developing our system , we discovered that an EBOV GP-induced fusion pore that connects two plasma membranes does not readily enlarge over time , in contrast to the pores of other viral fusion proteins . This anomalous lack of growth may be the reason cell-cell fusion has not been successfully observed in many prior attempts that used less sensitive assays to detect fusion . On the question of low pH , we found that activation of cathepsins by acidity is the sole cause for augmentation of fusion: if EBOV GP on the cell surface is artificially cleaved by thermolysin in the presence of cathepsin inhibitors , the extent of fusion is independent of pH .
We utilize fluorescent dye spread assays to monitor cell-cell fusion . Effector COS7 cells transfected to express EBOV GP were loaded with calcein-AM ( CaAM , green ) and pretreated with thermolysin ( Th ) and . It has been shown that thermolysin treatment cleaves GP1 on the viral membrane into a fusion-competent 18–19 kDa subunit [9 , 23 , 24] . Within the laboratory , thermolysin is therefore often used in lieu of membrane-bound cathepsins to cleave the GP1 subunit into a fusion-competent form . The COS7 cells were bound to 293T target cells that were either unlabeled or , for purposes of microscopic identification , loaded with the aqueous dye CMAC ( blue ) . We lowered the external pH for 10 min at room temperature , reneutralized , raised temperature to 37°C , and monitored dye spread at various times . We tracked the transfer of calcein between cells to quantify the extents of fusion; CMAC was used solely to identify the target cells . The fraction of cells that were stained by both calcein and CMAC , 2 hr after a 10-min low pH pulse , was comparable for cell-cell fusion mediated by EBOV GP , by Jaagsiekte sheep retrovirus ( JSRV ) Env , and influenza A virus ( IAV ) hemagglutinin ( HA ) —all requiring low pH for fusion to proceed ( Fig 1 ) . Fusion did not occur for effector cells that were mock transfected , establishing that CaAM transferred only due to fusion ( top row ) . It is often thought that EBOV fusion requires acidic pH [7 , 25 , 27] . But we found that thermolysin-treated effector cells expressing EBOV GP also fused to target cells at neutral pH ( 7 . 2 ) ( Fig 2A , bar 2 ) , albeit to a smaller extent than occurred 2 hrs after a 10 min exposure to an acidic pH of 5 . 7 at room temperature ( bar 1 ) . Representative images for dye transfer are shown to the right of the bar graphs ( Fig 2A ) . In its natural cellular setting , EBOV GP is cleaved not by thermolysin but by endosomal cathepsins B and L . In measuring fusion without prior thermolysin treatment of effector cells , we found that fusion still occurred , albeit to smaller extents ( Fig 2A , bars 3 and 4 ) . Again , a 10-min acidification ( Fig 2A , bar 3 ) led to greater amounts of fusion than occurred at neutral pH when measured after a 2-h reneutralization ( Fig 2A , bar 4 ) . Mock transfected effector cells , with or without thermolysin treatment , did not support any dye transfer at neutral or low pH , verifying that fusion required EBOV GP ( e . g . , see Fig 1 ) . The observed differences in extents of fusion between cells that were treated with thermolysin and those not were eliminated by long times of incubation after reneutralization ( Fig 2B ) . When EBOV GP was not cleaved by thermolysin , there was a 30 min latency between the fusion trigger ( acidification and raising temperature from 10°C to 37°C , Fig 2B , dark yellow circles ) and the occurrence of fusion . There was no latency when thermolysin cleaved the protein ( dark red squares , same fusion trigger as for dark yellow circles ) , suggesting that the 30 min latency when thermolysin was not used was due to the time it takes for a sufficient number of copies of cleaved GP to accumulate at a potential fusion site . The extent of fusion for non-treated effector cells ( dark yellow circles ) 2 hrs after reneutralization was almost equal to that observed after a 1 hr reneutralization for thermolysin-treated cells . But 4 hrs after a pH 5 . 7 pulse , the extent of fusion was independent of whether EBOV GP was cleaved by thermolysin . The kinetic difference is , to a large extent , likely due to the ~30 min latency until fusion occurs . The slopes of the linear portion for rates of fusion are comparable , suggesting that , after the latency , the kinetics of fusion are the same at low and neutral pH . The latency for EBOV GP-mediated fusion is much longer than for some viral fusion proteins , such as IAV HA [28] , but comparable to others , such as HIV Env in some studies [29] . We thus tested , at various times , whether some of the cell pairs that had not yet fully fused had hemifused: the addition of 0 . 5 mM CPZ to cell pairs ruptures hemifusion diaphragms that have formed between cell pairs , and this is a standard means to test for hemifusion [30 , 31 , 32] . We used thermolysin-treated effector cells to maximize cleavage of EBOV GP and found that adding CPZ either 30 , 45 , or 60 min after reneutralization did not induce any dye spread above that already observed , strongly indicating that a negligible percentage of cells were hemifused , but not fused , at any given time . NPC1 is an intracellular receptor for EBOV GP [14 , 33] . We compared extents of fusion for target parental HEK 293T cells versus target HEK 293T cells that stably overexpressed NPC1 . Effector cells that were not treated with thermolysin yielded fusion at pH 7 . 2 ( Fig 3A , bar 2 ) , and a greater extent of fusion after a 10-min acidic pH 5 . 7 pulse ( Fig 3A , bar 1 ) . The extent of calcein spread was greater for target cells overexpressing NPC1 ( Fig 3A , bars 3 and 4 ) than for parental 293T cells ( Fig 3A , bars 1 and 2 ) for matching conditions . Fusion was still pH-dependent for target cells overexpressing NPC1: calcein spread was greater 2 hr after a 10-min pH 5 . 7 pulse ( Fig 3A , bar 3 ) than in the absence of the pulse ( Fig 3A , bar 4 ) . We confirmed that fusion was dependent on the presence of NPC1 by generating and purifying a recombinant soluble protein consisting of domain C of NPC1 fused to GST ( denoted sNPC1 ) . The purity and size of sNPC1 was confirmed ( Fig 3B , inset ) . We added sNPC1 to the external solution and found that the extent of fusion increased monotonically with the amount of sNPC1 added ( Fig 3B ) , in accord with the prior demonstration that by binding NPC1 , EBOV GP undergoes conformational changes favorable for fusion [18] . The augmentation of fusion by sNPC1 indicated that , although there was a sufficient amount of NPC1 on cell surfaces to stimulate fusion , this amount was relatively small and fusion was consequently limited . NPC1 is an endosomal protein [34] , but a small fraction of NPC1 may be present on the plasma membrane of a cell . We assessed this possibility by using flow cytometry to measure binding with an antibody against NPC1 ( from LifeSpan Biosciences ) on parental 293 cells; shRNA that targeted NPC1 was stably expressed in one set of these 293 cells , and NPC1 was overexpressed in another set ( Fig 3C and 3D ) . The level of binding of the secondary FITC-labeled antibody against endogenous NPC1 ( as measured by mean fluorescence intensity , MFI ) was 3-fold greater than in the absence of the primary Ab ( Fig 3C , bar 1 vs . bar 4 , and Fig 3D ) . Expression was reduced for cells in which NPC1 was knocked down by shRNA ( bar 2 ) , and was greater for cells in which NPC1 was overexpressed ( bar 3 ) . These results demonstrate that copies of NPC1 reside in the plasma membrane of the 293 cells we used as targets in cell-cell fusion experiments . EBOV GP is certainly cleaved within endosomes as part of viral infection [26] . Because we observed cell-cell fusion at acidic pH without adding thermolysin , it is extremely likely that a fraction of GP on the cell surface was cleaved into a fusion-competent form . An antibody that only recognizes cleaved GP has not been reported , so we had to devise an alternate means to quantitatively measure the extent of cleavage . We were able to distinguish between the two forms of GP by using the property that NPC1 binds to cleaved , but not uncleaved , EBOV GP . We used a sNPC1 to examine cleaved GP by flow cytometry; in parallel , we measured the total amount of GP on cell surfaces by using an anti-FLAG antibody that bound to the FLAG tag on our GP construct . We also created a GP construct that was intrinsically more likely to be cleaved on the cell surface: we inserted the furin recognition site RRKR at amino acids 203–206 of GP1 ( referred to as GPfurin ) , the putative cleavage site for CatL in GP1 [16 , 35] . We reasoned that because exogenous expression of furin facilitates cleavage at this inserted site , generating the fusion-active 18–19 kDa subunit , the extent of cleavage of GP on the plasma membrane as well as the extent of cell-cell fusion would be greater for this construct than for WT . We experimentally confirmed our expectations: We determined the amount of cleaved GP on the cell surface by adding sNPC1 ( fused with GST ) to cells expressing either EBOV GP or GPfurin , and measuring their binding to an anti-GST antibody . This antibody was detected by a FITC-conjugated secondary antibody ( Fig 4 ) . The fraction of WT GP cleaved on parental cells ( Fig 4A , bar 1 ) was the same for cells that were transfected with both GP and furin ( bar 2 ) . The specificities of sNPC1 and antibody binding were confirmed by the 4–5 fold higher fluorescence than was seen for cells that did not express GP ( bar 5 ) . It is notable that cotransfection of cells by GPfurin and furin resulted in greater cleavage ( bar 4 vs bar 3 ) . We found that the expression of total WT GP as measured by the anti-FLAG antibody was not significantly altered by coexpression of furin ( Fig 4B , columns 1 and 2 ) , but cells that coexpressed GPfurin and furin consistently showed a decreased total GP ( compare bar 3 and 4 ) , possibly due to non-specific degradation of GPfurin . To determine the relative percentage of cleaved GP , we normalized cleaved GP by total GP . ( These are relative and not absolute percentages because different antibodies were used to detect cleaved vs . total GP . ) We found that a higher percentage of GP on the plasma membrane was cleaved for cells coexpressing GPfurin and furin than for cells expressing WT GP or GPfurin alone ( Fig 4C ) . Western blot analyses , using an anti-FLAG or an anti-GP1 antibody ( kind gift of James Cunningham ) , showed that the addition of furin increased the amount of cleaved GPfurin construct as compared to GPfurin alone ( Fig 4D , lanes 4 and 5 in left and right panels ) . Furin did not cleave any WT GP ( lanes 1 ) . We used these constructs to verify that an increased cleavage of EBOV GP led to a greater extent of fusion ( Fig 4E ) . Cotransfecting cells with GP and furin ( bar 2 ) led to the same extent of fusion as did transfection of GP alone ( bar 1 ) . In contrast , cotransfecting with GPfurin and furin led to more fusion ( bar 4 ) than transfecting only GPfurin ( bar 3 ) . Control experiments of transfecting only furin showed that furin , per se , did not promote fusion ( bar 5 ) . These experiments , taken together , establish that EBOV GP does appear on the cell surface , that some of it is cleaved , and that for the GPfurin construct , cleavage is augmented by coexpression of furin . To further confirm that the observed fusion was indeed mediated by EBOV GP , we utilized mutations that had previously been shown to greatly reduce viral infection [36] . We used MLV pseudovirus expressing GP , and observed that , indeed , the level of infection caused by the point mutant W597A ( Fig 5 , bar 2 ) , the double mutant G598A/G599A ( bar3 ) , and the point mutant I610A ( bar 4 ) were all substantially less than for WT GP ( Bar 1 ) . We then measured the extents of cell-cell fusion mediated by each of the mutant proteins . The extent of fusion in absence of thermolysin treatment supported by all three of the mutants ( Fig 5B , bars 2 , 3 , and 4 ) was much less than for WT GP ( bar 1 ) . Flow cytometry measurements , using the same cells as for fusion experiments , showed that each of the mutant GPs was well expressed on the cell surface ( Fig 5C ) . These experiments provide support that reduced infectivity by EBOV correlates with reduced GP-mediated fusion . We next tested 3 . 47 , a small molecule inhibitor against NPC1 , which prevents EBOV entry , as well as testing a neutralizing antibody ( KZ52 ) against EBOV GP . We found that both significantly reduced EBOV GP-mediated cell-cell fusion ( Fig 6A and 6B ) . The inhibitor 3 . 47 greatly reduced EBOV GP-mediated fusion but did not significantly alter cell-cell fusion induced by either Semliki Forest Virus ( SFV ) E1/E2 or IAV HA ( Fig 6A , 3 . 47 at 1 μM ) . Similarly , the neutralizing antibody KZ52 , which recognizes the interface between GP1 and GP2 [37] , reduced EBOV GP-mediated fusion , but not SFV-E1/E2 or IAV HA-induced fusion ( Fig 6B , KZ52 at 5 μg/ml ) . Higher concentrations of 3 . 47 completely inhibited fusion ( S1A Fig ) , but fusion was not further reduced by increasing the concentration of KZ52 beyond that employed in Fig 6B ( S1B Fig ) . Another central fingerprint of GP-mediated fusion is inhibition of EBOV infectivity by Bafilomycin A1 ( BafA1 ) . By neutralizing endosomes , BafA1 inhibits infection , at least in part , by reducing cathepsin activity which in turn results is reduced cleavage of GP1 . We found that addition of BafA1 ( 25 or 100 nM ) reduced the amount of cleaved GP that appeared on the cell surface ( Fig 6C , bar 2 vs bar 1 ) . This occurred despite a consistently greater amount of total GP in the plasma membrane after the addition of BafA1 ( Fig 6D ) . ( This greater amount was unexpected . Possibly , BafA1 prevented lysosomal degradation of GP . ) Normalizing the amount of cleaved GP by the total shows that cleavage of cell surface GP was significantly reduced by BafA1 ( Fig 6E ) . Thus , all data support the conclusion that the aqueous dye spread we observe is due to fusion induced by EBOV GP . Many of the effects of pH on kinetics and extents of EBOV GP-induced fusion we found were unexpected and quite different than those of pH-dependent fusion for other viral proteins . Notably , the extents of fusion did not monotonically increase as pH was progressively lowered , and the apparent pH dependence qualitatively varied with the times of reneutralization ( Fig 4 ) . After a pH 5 . 7 pulse , the extents of fusion were always greater than those achieved after more acidic pulses; following a pH 5 . 7 pulse ( at short incubation times ( i . e . , 30 min ) after the shift to neutral pH ) , more fusion was observed than for a less acidic pulse ( Fig 7A ) . However , for pH pulses of 5 . 7 and above , as the reneutralization time was increased , the extents of fusion became less dependent on pH; fusion was independent of pH for 5 . 7 and above after a 1 h reneutralization ( Fig 7A ) . In contrast , effector cells expressing IAV HA showed the typical and expected response of greater extents of fusion for lower pH values at all times after reneutralization; fusion events reached their full extents after a 30 min reneutralization ( Fig 7B , using the same protocol as for EBOV GP experiments ) . Thus , IAV HA induces fusion more rapidly than does EBOV GP . In separate experiments , we compared extents of EBOV GP-mediated fusion after a 4 h and 1 h reneutralization that followed 10 min , room temperature , acidic pH pulses ( Fig 7C ) . After the 4 h reneutralization , fusion was relatively independent of the acidity of the pH pulse , and a low pH pulse did not greatly augment fusion ( compare filled bars to open bars , Fig 7C ) . Equality in final extents of fusion at pH 5 . 7 and 7 . 2 could be a consequence of all cell pairs quickly fusing at low pH , thereby eliminating the possibility of further fusion , although we consider this unlikely . In addition to single cell measurements of aqueous dye transfer , we also monitored lipid dye continuity between effector cells ( treated with thermolysin ) and target cells . We labeled effector cells with the lipophilic fluorescent dye DiO and labeled target cells with DiI and determined extents of fusion by flow cytometry ( FACS ) . The double positive cells ( i . e . , the third quadrant ) are clearly products of hemifusion or cell-cell fusion ( Fig 7D ) . For effector cells treated with thermolysin , the percentage of fusion for the representative experiment was 18 . 5% at pH 5 . 7 , the optimal pH for fusion ( Fig 7D , second panel ) and only 1 . 5% at pH 5 . 0 ( first panel ) . Averaging six separate experiments for each condition , after a 2-h reneutralization , lipid mixing was greatest for a 10-min pH 5 . 7 pulse , and less for a pH 5 . 0 pulse than for cells maintained at neutral pH ( Fig 7E ) . The approximately two-fold greater fusion determined by flow cytometry at pH 5 . 7 than at 7 . 2 is also in agreement with the data for spread of calcein ( Fig 2 ) . For mock-transfected effector cells , virtually no lipid dye spread was observed between effector and target cells ( Fig 7E ) , in agreement with the aqueous dye spread measurements . Therefore it is clear that EBOV GP mediates a considerable amount of cell-cell fusion , and does so at an optimal pH of 5 . 7 . Once calcein movement from effector to target cell commenced , it continued for EBOV GP-mediated fusion , but at an extremely slow rate . In general , the fluorescence due to calcein never equalized between target and effector cells for EBOV GP-induced fusion ( Fig 8 ) . In contrast , for fusion pores created by other viral fusion proteins [33 , 38] , such as JSRV Env ( Fig 8A , upper images ) , the fluorescence did equalize . It is possible that the EBOV GP pores eventually closed , preventing calcein from attaining the same concentration in effector and target cells . We therefore quantified the rate of transfer of calcein by plotting calcein fluorescence of effector and target cells as a function of time ( Fig 8B ) . For EBOV GP-induced pores ( red curve ) , the transfer occurred over a time course of tens of minutes , and over this period the increasing fluorescence of a target cell never equalized the decreasing fluorescence of an effector cell ( Fig 8B ) . The fluorescence of the effector and target cells , on the other hand , equalized within a minute or so for JSRV Env-mediated pores ( Fig 8B , blue curve ) . The exceedingly slow transfer of calcein is a further indicator that EBOV GP-mediated pores remained extremely small . The fact that calcein transferred , albeit slowly over long times , shows that the fusion pores did not irreversibly close ( or if they did , new pores opened ) within tens of minutes of formation . As a control , we added saponin to effector cells and measured release of calcein to be sure that the dye did not become compartmentalized and therefore failed to transfer for reasons unrelated to the size of the fusion pore . Release was fast from the saponin-treated cells and was almost complete within 10 s , demonstrating that the overwhelming majority of calcein was , in fact , free and mobile . We further studied the size and rate of growth of EBOV GP-mediated fusion pores by assessing the size of dyes that can permeate these pores over time . We loaded effector cells with CMTMR in addition to calcein . CMTMR forms disulfide bonds with the tri-peptide glutathione , and these complexes are somewhat larger than calcein . The complexed glutathione can also form disulfide bonds with cytosolic proteins and hence CMTMR fluorescently labels proteins that are much larger than calcein . As a consequence , the size distribution of molecules labeled by CMTMR is expected to be quite diverse , some only somewhat larger than calcein and others very much larger . We found that CMTMR transferred for only 2–3% of the cell pairs for which calcein exchange occurred ( Fig 8C ) . The relative inability of CMTMR to spread indicates that fusion pores typically did not enlarge sufficiently to allow passage of a molecule of the size of the nucleocapsid of EBOV . In actual viral infection , factors absent in our model system are probably promoting expansion of the fusion pore connecting an envelope and endosomal membrane . We used electrical capacitance measurements to directly and quantitatively assess fusion pore size . The slow time course for EBOV GP-mediated fusion necessitated that the tight electrical seal between the patch pipette and plasma membrane be maintained for long times . This proved difficult in practice . We were able , however , to electrically observe pores between cell pairs in three cases , and in these cases the pores never enlarged within 30 s of formation and generally fluctuated within small values of conductance ( Fig 9A ) . The conductance of the fluctuating pores did not return to baseline , showing that the pores did not close , but instead remained restricted to a small size . By way of comparison , it can be readily seen from representative traces of electrically measured fusion pores created by other viral proteins ( Fig 9B ) that fusion pores generally significantly enlarge over time . The absence of pore enlargement for EBOV GP suggests that many of the prior attempts at monitoring cell-cell fusion mediated by this fusion protein did not succeed because the reporter molecules that needed to permeate the fusion pore for detection of fusion were too large to pass through the pore . Although only three pores were electrophysiologically measured , the finding that each of them did not exhibit increased conductance over time implies that the slow passage of fluorescent dyes through them was not due to structures that prevent their access to the pores . Slow pore enlargement could be due to a number of factors , including slow recruitment and incorporation of additional copies of cleaved GP into the wall of the pore , or slow accumulation of lipids into the wall . We added NH4Cl to external media to test whether acidic intracellular compartments were essential for EBOV GP-mediated cell-cell fusion . The addition of 10 mM NH4Cl greatly reduced fusion after a 10-min pH 5 . 7 pulse in the absence of thermolysin treatment , so as to avoid activating uncleaved EBOV GP on the cell surface ( Fig 10A ) . In contrast , the addition of 10 mM NH4Cl did not affect fusion induced by an optimal pH pulse for either SFV E1/E2 or IAV HA ( Fig 10A ) . Similarly , 100 μM chloroquine inhibited cell-cell fusion mediated by the fusion protein of EBOV , but not by the proteins from either SFV or IAV ( Fig 10A ) . The elimination of fusion by the addition of 10 mM NH4Cl ( bar 2; same conditions as in Fig 10A ) was most likely caused by reducing cathepsin activity through neutralization of intracellular compartments: it was largely reversed by adding a recombinant cathepsin B to the external solution ( Fig 10B , bar 3 ) . Because the normally acidic intracellular compartments were neutralized by NH4Cl , the pool of EBOV GP on the cell surface that was previously uncleaved must have been cleaved by the added membrane-impermeant recombinant protease . The dose-response curves for inhibition of fusion by chloroquine ( Fig 10C ) or NH4Cl ( Fig 10D ) verified that inhibition of fusion is increased with increasing concentration of the neutralizing agent . Therefore , even if the external solution is acidified , EBOV GP-mediated cell-cell fusion does not occur unless the acidity of intracellular organelles is maintained . We conclude that EBOV GP present on the cell surface requires an intracellular compartment for cleavage , as is consistent with previous reports . It is possible , however , that there are copies of cathepsins in the plasma membrane , and acidification of the external solution activates them to cleave EBOV GP . Proteinase K ( PK ) has proved useful for assessing conformational changes that viral proteins undergo at different stages of fusion [30 , 39] . We found that EBOV GP was PK-sensitive for all steps of the fusion process ( S1 Text and S2A Fig ) , that fusion was restored over time after removing PK ( S2B Fig ) , and that normal cellular trafficking of protein led to replacement of proteolytically digested GP with newly delivered intact GP ( S3 Fig ) . We also used Brefeldin A ( BFA , 50 μM ) —an inhibitor of trafficking from endoplasmic reticulum to Golgi—to further characterize the consequences for fusion of altering intracellular trafficking of EBOV GP . Here we found that treatments expected to increase the amounts of cleaved EBOV GP on the cell surface led to greater extents of fusion ( S1 Text and S4 Fig ) . For virus internalized in endosomes , EBOV GP is believed to be cleaved by cathepsins B and L , but not by cathepsins A or D . We prevented cathepsin-induced cleavage by treating bound effector and target cells with a cathepsin B inhibitor ( CA-074 ) or a cathepsin L inhibitor ( Z-FY-CHO ) . In the absence of thermolysin treatment , the inhibitors led to significantly reduced fusion at both neutral and low pH ( Fig 11A , compare “untreated” and “treated”: as always , changes of solutions containing membrane-impermeant buffers were used to control pH ) . Using inhibitors against cathepsin A ( lactacystin ) or cathepsin D ( pepstatin A ) –neither of which is thought to cleave EBOV GP–did not lead to reduced fusion using the same protocol as for the cathepsin B and L inhibition experiments ( Fig 11A ) . These results provide strong support that fusion observed in our experiments in the absence of thermolysin treatment is due to , at least in part , copies of EBOV GP on the cell surface that have their GP1 subunits cleaved by cathepsins . These results also document that neither cathepsin A or D cleaves EBOV into a fusion-competent form . From the results as a whole , it is clear that low pH does not induce fusion unless the GP1 subunit has been cleaved . It is known that cathepsin activity is increased by acidity . We suggest that low pH acts , at least in part , by augmenting cathepsin activity on the cell surface . The same pattern of pH-dependence of fusion was observed for effector cells treated with thermolysin while cathepsin activity was continually inhibited: fusion was dependent on pH and was significantly reduced by the cathepsin B inhibitor or the cathepsin L inhibitor ( Fig 11A , thermolysin-treated , bars 2 and 3 of each set of columns ) , but was relatively unaffected by cathepsin A or D inhibitors ( bars 4 and 5 ) . Cell-cell fusion exhibits a maximum at pH 5 . 7 ( column 4 compared to column 1–3 ) . Several cathepsins exhibit maximal activity in the pH range of 5 . 5 to 6 . 8 [40] , so the maximum extent of fusion at pH 5 . 7 would likely be due to the pH dependence of cathepsin activity on the cell surface . Control experiments provide additional support for the conclusion that cathepsins aid EBOV GP-mediated fusion between cells . Blocking cathepsin B ( by adding the cathepsin B inhibitor ) immediately after application of an acidic pH pulse resulted in a substantial reduction in the extent of fusion after a 2-h reneutralization ( Fig 11B , effector cells were thermolysin-treated ) . The reduction from the control was ~2-fold; a 2-fold reduction also occurred when the cathepsin B inhibitor was constantly present ( see Fig 12B ) . The nearly equal percentages of inhibition of fusion are expected , since in the presence of the cathepsin inhibitor , uncleaved copies of EBOV GP would not be cleaved during the period of reneutralization . Thus , low pH appears to promote cleavage of EBOV GP by cathepsins on the cell surface . Incubating effector cells that were not treated with thermolysin with a recombinant human cathepsin B ( rhCat B ) ( Fig 11C , bar 2 ) increased fusion significantly over the control ( bar 1 ) . The simplest explanation for this increase is that the recombinant protein led to a higher level of GP1 cleavage than that induced by endogenous cellular cathepsins . To explicitly test whether increasing the activity of cathepsin increased the likelihood that GP on the cell surface was cleaved , we cotransfected cells to express GP and cathepsin B and used sNPC1 to measure the percentage of GP in the plasma membrane that was cleaved ( as described for Fig 4 ) . This percentage was greater ( Fig 11 , column 2 ) than the control ( column 1 ) in the presence of cathepsin B transfection . Using the same techniques , we also showed that adding thermolysin to solution did indeed increase cleavage of cell surface GP ( Fig 11D ) . Does low pH directly cause conformational changes in EBOV GP to induce fusion , or does it work via increasing the activity of cathepsins , or both [7 , 20] ? We were able to approach these questions by using the ability of BFA to effectively block delivery of EBOV GP to the cell surface and , independently , by using cathepsin inhibitors to prevent GP cleavage . We incubated effector cells with BFA for 45 min to prevent further delivery of EBOV GP to the plasma membrane prior to a thermolysin-treatment , and maintained the presence of the drug during all solution changes . The extent of fusion was independent of pH , and considerably less than when the trafficking inhibitor was not employed ( Fig 12A ) . The clear conclusion is that , with BFA present , all fusion was caused by copies of EBOV GP that had been cleaved by thermolysin and that remained on the cell surface . The finding that pH pulses did not affect the extent of fusion at all shows that acidity did not promote the conformational changes in cleaved EBOV GP that would lead to fusion . We inhibited cathepsin activity to further test the conclusion that once cleaved , EBOV GP no longer requires low pH to induce fusion . We performed experiments in which CA-074 , a cathepsin B inhibitor , was continually present ( Fig 12B ) . The inhibitor was added to isolated effector cells and maintained for 45 min to ensure that all EBOV GP delivered to the plasma membrane was not cleaved . Effector cells were then thermolysin-treated , always maintaining the inhibitor . These cells were bound to target cells , and the external solution was acidified to pH 5 . 7; after reneutralization , cells were maintained for 1 h at 37°C , with all manipulations performed in the presence of the inhibitor ( Fig 12B , experimental protocol illustrated on top ) . The extent of fusion was greater when the inhibitor was not added ( control ) : this indicates that delivery to the cell surface of EBOV GP cleaved by endosomal cathepsin ( subsequent to thermolysin treatment ) significantly contributes to fusion . More importantly—and central to the mechanism of EBOV GP-mediated fusion—the extent of fusion was independent of pH . This finding strongly implies that acidic conditions have no direct effect on EBOV GP-mediated fusion . The pH dependence of fusion is solely due to the ability of cathepsin to cleave EBOV GP; once cleaved , acidic conditions directly induce conformational changes in cleaved EBOV GP that lead to fusion .
Using both cytoplasmic and membrane dye transfer assays , we established that known properties of EBOV fusion occurring within endosomes are replicated by our cell-cell fusion system and that specific inhibitors of EBOV infection—the small molecule inhibitor 3 . 47 and a neutralizing antibody KZ52—block fusion . The inhibition of EBOV GP-mediated cell-cell fusion ( but not IAV HA or SFV E1/E2 fusion ) by the lysosomotropic agents NH4Cl and chloroquine is expected: EBOV GP cleavage is eliminated because cathepsin activity is greatly reduced by neutralization of endosomes; inhibiting cathepsin activity reduces cleavage of EBOV GP [41 , 42] . We also showed that copies of NPC1 reside in the target membrane and some GP resides in the plasma membrane , and that a fraction of the GP is properly cleaved . It is virtually certain that the cell-cell fusion process investigated in the present study is mediated by EBOV GP . It is likely that past lack of success in observing cell-cell fusion is attributable to the fact that the fusion pore mediated by EBOV GP on the cell surface remains small . Over the time scales of electrical measurements , the pore does not enlarge at all . Based on fluorescence dye spread measurements , it enlarges more slowly and to a lesser extent than any other pore mediated by a viral fusion protein of which we are aware , and it may even tend to close . The EBOV GP fusion pore is large enough to allow the passage of calcein , but just barely . There is little doubt that the fluorescent dye CMTMR does not permeate the pore because virtually all of it complexes with proteins; the complex becomes permeable only after a pore enlarges . Electrical measurements directly demonstrate that the EBOV GP-induced pore remains small . Over the course of time in our cell-cell fusion experiments , EBOV GP-mediated fusion pores do not significantly enlarge . The question now becomes: how readily does a fusion pore enlarge when connecting an EBOV envelope with an endosomal membrane ? This fusion pore must expand to sizes that permit passage of the large viral nucleocapsid . Four major possibilities present themselves: ( i ) the necessary enlargement is extremely slow for the endosome-viral pores; ( ii ) elements engaging plasma but not endosomal membranes , such as cytoskeleton , retard the growth of fusion pores; ( iii ) a protein ( such as the two-pore calcium channel , present in the endosomal compartments that support EBOV fusion [43] ) is required for pore enlargement; ( iv ) control of calcium concentrations ( e . g . , through the two-pore channels ) regulates fusion pore formation or enlargement in endosomes . Methods to monitor the formation and growth of fusion pores of EBOV GP-bearing viral particles within endosomes will be needed to answer these questions [44] . We have unambiguously shown that a fraction of EBOV GP on the cell surface is cleaved . By using the GPfurin construct we also demonstrated that increased cleavage correlates with greater fusion . In addition , we functionally evaluated the cleavage status of EBOV GP on the cell surface by adding a water-soluble recombinant cathepsin B or thermolysin to solution and found that these proteases promoted fusion . Late endosomes and lysosomes are generally thought to be the cellular site of cleavage of EBOV GP by cathepsins [45]; it is likely that a fraction of EBOV GP is cleaved within endosomes and then recycled to the plasma membrane where it mediates cell-cell fusion , independent of pH ( Fig 12 ) . We suggest that uncleaved EBOV GP that reaches the surface is cleaved upon acidification of the external solution by cathepsins within the plasma membrane . Thermolysin cleaves uncleaved copies of EBOV GP that are delivered to the cell surface , accounting for the enhancement of fusion by the addition of the protease . Regardless of the site of GP cleavage , an appreciable fraction of the GP1 subunit is indeed cleaved into its fusion-competent form after the addition of thermolysin . NPC1 serves as receptor for EBOV GP in endosomes , and is essential for the virus to infect a cell . We have now shown that NPC1 is not confined only to intracellular membranes , but rather that some copies reside in plasma membranes . Our finding that sNPC1 promotes EBOV GP-mediated cell-cell fusion suggests that domain C of NPC1 alone is sufficient to induce the needed conformational changes in the fusion protein . It is well established that the extents of cell-cell fusion correlate with the levels of viral fusion protein expression on cell surfaces [46 , 47] . Thus , it is not surprising that the extents of cell-cell fusion induced by EBOV GP are affected by its delivery to , and loss from , plasma membranes . For some viral fusion proteins , such as the paramyxovirus Hendra and Nipah virus F proteins , and SARS coronavirus S protein , cell-cell fusion is sensitive to protein cycling [48 , 49 , 50] . These proteins require acidic intracellular compartments for cleavage: endosomes for Nipah virus [51] , and endosomes and the Trans-Golgi Network for Hendra virus [52] . But the dependence of cell-cell fusion on protein trafficking is unusual for typical pH-dependent viral fusion proteins; for these proteins acidity does not promote cleavage , but instead directly induces conformational changes [53] . Once activated , these typical low-pH-dependent fusion proteins quickly inactivate if they do not promote fusion [54] . Hence , protein delivered to the cell surface subsequent to an acidic pulse will not be able to promote fusion . In contrast , proteins that induce fusion at neutral pH will promote fusion once they are delivered to the cell surface . Our results show that EBOV GP cleaved by endosomal cathepsins are no longer sensitive to pH and therefore can induce fusion once they arrive at the plasma membrane . Infectivity is subject to processes other than fusion , and so infectivity need not always correlate with extents of cell-cell fusion . For example , it has recently been shown that the activity of two-pore calcium channels in endosomes is required for EBOV infection [43] , and that EBOV infects by fusing to endosomal membranes that contain both NPC1 and the two-pore channel [44] . Tetrandrine blocks these channels and inhibits EBOV infection . We found that tetrandrine ( 150 nM ) did not affect EBOV GP-mediated cell-cell fusion . It is notable that the extent of fusion that occurs after 4 h at neutral pH was roughly equal to the extent that follows a pH 5 . 7 pulse . Because fusion induced by cleaved GP is pH-independent , we interpret the continual increase in fusion at neutral pH over time to be a consequence of intracellular trafficking: new copies of EBOV GP continually replace or supplement old copies and these new/supplemented , cleaved copies can cause fusion between cells that had not previously fused . Thus , acidification likely promotes more fusion at early times through activation of surface cathepsins that cleave EBOV GP . However , it is not presently clear why fusion kinetics is faster after a pH 5 . 7 pulse than after a pH 5 . 4 ( or more acidic ) pulse . The pH dependence of cathepsin activity is complicated [55] . While activity generally increases with acidification , some cathepsins exhibit the same activity in the range of pH 7 as at lower values of pH [56] . For others , activities are maximal at an intermediate pH , such as 5 . 7 [40] . Also , the pH dependence of cathepsin activity varies with environment and conditions , such as redox potentials on each side of the membrane in which a cathepsin resides [57 , 58] . Any relevant cathepsins ( e . g . , B or L ) on the cell surface can , at their optimal pH , cleave EBOV GP . A direct test of whether EBOV GP on the cell surface is maximally cleaved by cathepsins at pH 5 . 7 will require methods to measure the percentage of EBOV GP that is cleaved as well as cathepsin activity at the cell surface . The role of acidic pH in EBOV fusion has been debated in the field [7 , 20 , 59] . Our data unambiguously show that cell-cell fusion is regulated by extracellular pH . The acidity of the extracellular solution can , in principle , augment both the activity of cathepsins that reside in the plasma membrane and directly promote conformational changes of cleaved EBOV GP on the cell surface . ( Although cathepsins are regarded as endosomal membrane proteins , some copies also likely reside in plasma membranes from which many endosomes derive [60] ) . The great reductions in fusion caused by inhibition of cathepsin activity and recovery of fusion by addition of a recombinant cathepsin establish that cathepsins’ activity at the cell surface is consequential . Independent experiments in which cathepsin activity was inhibited or delivery of protein to cell surface was blocked show that the pH-dependence of fusion is eliminated once EBOV GP is cleaved . This demonstrates that fusion mediated by the cleaved form is intrinsically pH-independent . That is , cleaved EBOV GP is essentially a neutral pH fusion-inducing protein; all the experimentally observed and biological relevant pH-dependence is a consequence of cathepsin activity . The faster kinetics of cell-cell fusion after a pH 5 . 7 pulse than for pulses at higher values can be accounted for by greater cleavage of EBOV at the cells surface at pH 5 . 7 . A previous study used model peptides to mimic the six-helix bundle of EBOV GP2 and found that low pH increased bundle stability [61] . The stage of fusion in which bundle formation occurs has not been identified for EBOV GP . It may be , for example , that the bundles form subsequent to pore formation , as occurs for HIV Env [29] , and that increased bundle stability aids pore enlargement , but not fusion itself [62] . Alternatively , the model peptide may not mimic bundle stability within a full length , structurally intact GP . As a general rule , fusion kinetics for viral proteins that induce fusion at neutral pH are slower than for proteins that utilize low pH as a trigger . This could explain the slow fusion kinetics of EBOV GP mediated fusion , despite classification as a low pH-requiring process . Alternatively , a need to continually deliver EBOV GP could be the reason EBOV GP-mediated cell-cell fusion is slow . The development of an experimentally convenient system of EBOV GP-mediated fusion should make it possible to determine molecular mechanisms by which EBOV releases its genome into infected cells . Our results and conclusions are diagrammatically summarized in Fig 13 . NPC1 is an intracellular receptor for EBOV GP within endosomes . But , as we have shown , NPC1 can also reside in the plasma membrane . Endosomal cathepsins cleave EBOV GP , and any cathepsins that reside in plasma membranes will also cleave surface GP upon acidification of the external solution . Both cleaved and uncleaved copies of EBOV GP are continually delivered to and retrieved from the surface , and hence intracellular trafficking contributes to extents and kinetics of fusion . But binding of EBOV GP to the target membrane should inhibit endocytotic retrieval . Consequently , EBOV GP ( both cleaved and uncleaved ) should accumulate at potential fusion sites , leading to more fusion over time . Preventing acidification of endosomes to block cleavage of EBOV GP , or inhibiting delivery of the protein to the cell surface , greatly reduces fusion . Acidification of the external solution to pH 5 . 7 increases the activity of cathepsins that reside in the cell surface , and this results in additional cleavage of EBOV GP . The addition of thermolysin converts all surface GP to the cleaved form , thereby resulting in the maximal extent of fusion . That cell-cell fusion induced by cleaved EBOV GP does not depend on pH , provides critical insight into the mechanism of EBOV entry and infection .
Purchased reagents were: Lactacystin ( a cathepsin A inhibitor , Santa Cruz Biotechnology , Dallas , TX ) ; pepstatin A ( a cathpesin D inhibitor , Santa Cruz Biotechnology ) ; Cathepsin L inhibitor ( catalog no . sc-3132 , Santa Cruz Biotechnology ) ; CA-074 ( Cathespin B inhibitor , Calbiochem ) ; Recombinant human cathepsin B ( R &D Systems , Fisher Scientific ) , Brefeldin A ( Cayman Chemicals , Ann Arbor , MI ) , ) ; poly-lysine ( M . W . 70 kD , Sigma ) ; bovine serum albumin ( BSA , Sigma ) , Thermolysin ( Sigma ) ; Proteinase K , chlorpromazine ( CPZ ) ( Sigma ) , lentiviral shRNA targeting NPC1 ( Sigma ) , anti-NPC1 ( LifeSpan BioSciences , Seattle , WA ) , anti-FLAG and anti-ß-actin antibody ( Sigma ) ; anti-GP1 anitbody ( gift of James Cunningham , Harvard Medical School , Boston , MA ) . PBS++ and DMEM were obtained from Gibco ( Grand Island , NY ) . All fluorescent probes were purchased from Molecular Probes ( Life Technologies , Eugene , OR ) . The mucin-deleted EBOV GP construct was originally obtained from David Sanders ( Purdue University , West Lafayette , IN ) . For this work , we mainly used an N-terminal FLAG-tagged , mucin-deleted EBOV GP construct by replacing the signal peptide of GP with that of preprotrypsin followed by a FLAG sequence . All GP mutants , including GPfurin , were made by overlapping PCR using the FLAG-tagged GP construct as the template . The plasmid to express JSRV Env has been described [38] . To express IAV HA for dye spread experiments , we used the X31 strain ( plasmid provided by Judith White , University of Virginia , Charlottesville , VA ) . A standard calcium phosphate method was used to express SFV E1/E2 via transfection of the pCB3-wt vector , plasmid provided by Margaret Kielian , Albert Einstein College of Medicine , Bronx , NY [63] . A small molecular inhibitor , 3 . 47 , was a gift of James Cunningham ( Harvard Medical School , Boston , MA ) . HEK 293T and COS7 cells employed have been previously described [64] . The HEK 293T cells stably expressing were generated by transducing cells with pBabe retroviral vector expressing NPC1 ( gift of Kartik Chandran ) followed by puromycin selection ( Sigma , 2 μg/ml ) . All cells were grown in Dulbecco’s modified Eagle’s ( DMEM ) medium , supplemented with 0 . 5% penicillin/streptomycin plus 10% fetal bovine serum ( FBS ) . For all experiments using EBOV GP , COS7 cells were maintained in Eagle’s Medium with glucose , L-glutamine , and sodium pyruvate , supplemented with 10% Cosmic Calf Serum ( HyClone , Logan , Utah ) , Pen Strep ( Gibco ) , and 0 . 5 mg/ml G418 Sulfate ( Cell Gro , Manassas , VA ) , and transfected to express EBOV GP by a standard calcium phosphate procedure [65] . About ~ 2x106 cells were loaded with 1 . 5 μM calcein-AM as previously described [29] and sometimes coloaded with 1 μM 5- ( and-6 ) - ( ( ( 4-chloromethyl ) benzoyl ) amino ) tetramethylrhodamine ) ( CMTMR ) [29] . If these effector cells were thermolysin-treated to cleave EBOV GP , 200 μg/ml thermolysin was incubated with the cells for 20 min at room temperature . Exchanging the solution with DMEM removed thermolysin; residual thermolysin was further removed by spinning down the cells three times and replacing the aqueous solution . HEK 293T cells were maintained in the same media and in the same way as COS7 cells and were used as targets . ~ 2x106 cells were loaded with 20 μM CMAC . Effector cells were mixed , including a gentle vortex—in a tube containing either PBS++ ( sometimes supplemented with 1 mg/ml BSA ) or DMEM—with the labeled target HEK 293T cells . The cells were added into polylysine-coated ( 1 mg/ml ) wells of an 8-well slide ( Thermo Fisher ) [29] and allowed to settle and adhere to the bottom for 30 min at room temperature . The pH was lowered ( or not ) for 10 min at room temperature to the indicated value ( pH 5 . 7 unless stated otherwise , using an exchange solution consisting of 100 mM NaCl , 1 . 5 mM KCl , 2 . 5 mM MgCl2 , 2 . 5 mM CaCl2 , 20 mM MES ) , the solution was then reneutralized to pH 7 . 2 by an exchange of solutions , and the temperature raised to 37°C . After this reneutralization for the indicated time , generally 2 h , fusion was scored as a function of time by the transfer of calcein into target cells , as described [64] . For fusion experiments utilizing IVA HA as control , the expressed HA in effector cells were was cleaved into HA1-HA2 subunits with trypsin , as previously described [28] . To label effector cells , ~ 2x106 cells/ml were incubated with 10 μM DiO for 30 min at 37°C . The day before an experiment , target cells were split and plated on glass cover slips placed in culture dishes so as to allow convenient transfer . These target cells were labeled by 100 μM DiI for 30 min at 37°C . Labeled effector cells were thermolysin treated ( 200 μg/ml ) and added above labeled target cells . Binding was allowed to occur for 40 min at room temperature before washing out unbound effector cells . The solutions bathing the cover slips ( one cover slip per culture dish ) was acidified to the indicated pH for 10 min at room temperature , and the culture dish placed in a 37°C incubator for 2 h . Cells were detached from cover slips by adding 10 μg/ml trypsin to a divalent-free solution containing 0 . 5 mM EDTA for 10 min at room temperature , followed by vigorous , repeated pipetting to dissociate bound ( i . e . , neither hemifused or fused ) cells . Colocalization of the two lipid dyes was monitored by flow cytometry ( Guava Easy Cite , Guava Technology , Millipore ) , using two channels emission , one for each dye ( 515 nm for DiO and 560 nm for DiI; both excited by a 488 nm laser ) . The same protocols were followed for mock-transfected COS7 cells; this data was subtracted from data obtain for COS7 cells expressing EBOV GP to obtain percentages of fused cells . The domain C of NPC1 was cloned into a pGEX-4T1 vector that had a GST tag on the N-terminus ( GE Healthcare Life Sciences , Pittsburgh , PA ) . The expression of fusion protein was induced in E . Coli . by IPTG ( 0 . 5 mM ) and purified by glutathione sepharose 4B ( GE Healthcare Life Sciences ) . Protein was quantified by a Bradford assay and used for cell-cell fusion and for measurements of cleaved GP . The expression of NPC1 on 293T cell surfaces was determined by using anti-NPC1 ( against N-terminus 34–174 aa; LifeSpan BioSciences ) . Determination of EBOV GP cleavage . HEK293T cells were transfected with EBOV GP or GPfurin in the presence or absence of a plasmid that encodes furin ( kind gift of Paul Bates , University of Pennsylvania ) . Transfected cells were detached by PBS containing 5 mM EDTA . One portion ( 1 million cells ) was used to measure the cleaved GP by incubating cells with 2 μg sNPC1 for 2 h on ice , followed by adding a mouse monoclonal anti-GST antibody; the fluorescence signal was quantified by adding a FITC-conjugated secondary anti-mouse antibody using flow cytometry . Another portion of the transfected cells ( also 1 million ) was used to measure the total GP expression on the cell surface , using an anti-FLAG antibody . The total GP , as determined by mean fluorescence intensity ( MFI ) , was used to determine the percentage of GP on the plasma membrane that was cleaved: the amount of cleaved normalized by total GP . Cleavage of GPfurin in cell lysates was determined by Western blotting using anti-FLAG or an anti-GP1 antibody . Production of MLV retroviral pseudotypes bearing EBOV GP and viral infection were as described previously [66] . Briefly , 293 GP/LAPSN packaging cells stably expressing MLV Gag-Pol and alkaline phosphatase ( AP ) were transfected with plasmids encoding the EBOV GP or mutants , and the viruses produced were used to infect HTX cells ( a subclone of HT10180 ) . Viral infectivity was determined by counting AP+ foci 72 h after infection . We proteolytically determined the stages of fusion at which EBOV GP was proteinase K-sensitive by treating cells with 0 . 2 mg/ml proteinase K for 20 min , and maintaining or removing it as indicated , at various points in our protocol ( S1 Fig ) . We cleaved EBOV GP with thermolysin and used our standard protocol ( a 10-min pH 5 . 7 pulse ) , incubated the cells for 2 h at pH 7 . 2 , and then measured fusion . All experiments were performed in parallel on the same days; as controls , proteinase K was not employed and extents of fusion were measured . Brefeldin A ( 50 μM ) was used to inhibit anterograde protein trafficking as described in the experiments of S4 Fig . We determined the latency between lowering pH and fusion by using cells mixed within a tube , placed over poly-lysine coated cover slips within a culture dish maintained at 10°C and allowed to settle for 30 min . The pH was lowered to the indicated value for 10 min at room temperature through an exchange of solutions . The cover slips were then transferred into dishes at 37°C , neutral pH , for indicated times . The time of transfer is defined as time = 0 . For thermolysin-treated cells , the effector cells were treated prior to binding to target cells . The extents of fusion were quantified at varied times by spread of calcein . We used the rate of accumulation of calcein into the target cell and its depletion in effector cells to access the size of the fusion pore as a function of time [62] . The fluorescence of both effector and target cells was proportional to calcein concentration , as verified by the procedure detailed in [62] . NH4Cl ( 10 mM unless otherwise noted ) was added to the bathing solution and maintained throughout the course of an experiment , including during any low pH pulses , to obtain neutralization . The same procedure was used with chloroquine ( 100 μM unless stated otherwise ) or BafA1 ( 25 or 100 nM ) to cause endosome neutralization . The ionic and buffering contents of the NH4Cl-containing low pH solutions were pre-adjusted to the required pH so as to maintain osmotic strength at 290 mOsM during low pH pulses . To monitor EBOV GP expression and its recovery after proteinase K treatment , cells were treated with 0 . 2 mg/ml of the protease for 20 min at room temperature , and the staining protocol now described was used immediately or after cells were maintained for 3 h in DMEM at 37°C , as indicated . EBOV GP-expressing COS7 cells were transferred to 15 ml tubes and incubated for 1 h at room temperature with a primary antibody ( human anti-EBOV GP-KZ52 stock at 1 . 3 mg/ml , IBT Bioservices , Gaithersburg , MD ) that was diluted 1:200 in PBS++ that was supplemented with 10% fetal bovine serum . After three washes with the 10% FBS-PBS++ solutions , a secondary FITC-conjugated goat anti-human antibody ( Fisher Scientific ) was added at a final concentration of 100 μg/ml and maintained for 45 min at room temperature in the dark . After cells were washed twice , they were added to 8-well slides that had been treated with poly-L-lysine ( M . W . 70 , 000–150 , 000 , Sigma Aldrich , St . Louis , MO ) . The cells were then fixed for 20 min at room temperature with 2% paraformaldehyde , and washed twice with DMEM . Pore size over time was also quantified by using patch clamp time-resolved admittance measurements [64] , often referred to as capacitance measurements . Fusion was promoted by a 10-min pH 5 . 7 pulse at room temperature . Because the fusion pore took considerable time to form , and a seal between the patch pipette and cell could only be maintained once solutions and temperature were established , temperature was raised to 37°C for ~15 min before attempting electrical measurements . ( This procedure reduced the time between establishing the seal and fusion pore formation ) . The cover slip was then placed in a temperature-controlled chamber on a microscope stage , and the seal established . The external solution consisted of 135 mM N-methyl-glucamine aspartate-5 mM MgCl2-2 mM HEPES ( pH 7 . 2 ) ; the solution within the patch pipette was 135 mM cesium glutamate-5 mM MgCl2-5 mM BAPTA [1 , 2-bis ( o-aminophenoxy ) ethane-N , N , N′ , N′-tetraacetate]-10 mM HEPES ( pH 7 . 2 ) . To electrically characterize fusion pores created by IAV HA , ASLV Env , and HIV Env , we used cell lines that stably express the fusion protein and target cells that stably express the cognate receptor: HAb2 cells that express IAV HA as effectors and 293T cells as targets [64]; NIH 3T3 EnvA cells that express ASLV Env and 293T TVA cells as targets [64 , 67]; and TF228 cells that express HIV Env and Hela T4 cells as targets [29] . Pair-wise Student t-tests were used to compare the outcome of a manipulation on fusion as compared to the control . In figures , unless otherwise indicated , a single asterisk ( * ) denotes p < 0 . 05 , two asterisks ( ** ) denotes p < 0 . 01 , and three asterisks ( *** ) denotes p < 0 . 001 . | The devastation and transmissibility of Ebola virus ( EBOV ) are well known . However , the manner in which EBOV enters host cells through endosomal membrane remains elusive . Here , we have developed a convenient experimental system to mimic EBOV fusion in endosomes: cells expressing the fusion protein of EBOV , GP , on their surface are fused to target cells . This system exhibits the known key properties of EBOV fusion . We show that the pH-dependence of EBOV fusion is caused by the pH-dependence of cathepsins , proteases known to cleave EBOV GP into a fusion-competent form . We demonstrate that the fusion activity of this cleaved form is independent of pH . We further show that the enlargement of the fusion pore created by EBOV GP is unusually slow in reaching sizes necessary to pass EBOV’s genome—this is atypical of virally created fusion pores . This cell-cell fusion system should provide a useful platform for developing drugs against EBOV infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2016 | Induction of Cell-Cell Fusion by Ebola Virus Glycoprotein: Low pH Is Not a Trigger |
Gliomas , the most common malignant tumors of the nervous system , frequently harbor mutations that activate the epidermal growth factor receptor ( EGFR ) and phosphatidylinositol-3 kinase ( PI3K ) signaling pathways . To investigate the genetic basis of this disease , we developed a glioma model in Drosophila . We found that constitutive coactivation of EGFR-Ras and PI3K pathways in Drosophila glia and glial precursors gives rise to neoplastic , invasive glial cells that create transplantable tumor-like growths , mimicking human glioma . Our model represents a robust organotypic and cell-type-specific Drosophila cancer model in which malignant cells are created by mutations in signature genes and pathways thought to be driving forces in a homologous human cancer . Genetic analyses demonstrated that EGFR and PI3K initiate malignant neoplastic transformation via a combinatorial genetic network composed primarily of other pathways commonly mutated or activated in human glioma , including the Tor , Myc , G1 Cyclins-Cdks , and Rb-E2F pathways . This network acts synergistically to coordinately stimulate cell cycle entry and progression , protein translation , and inappropriate cellular growth and migration . In particular , we found that the fly orthologs of CyclinE , Cdc25 , and Myc are key rate-limiting genes required for glial neoplasia . Moreover , orthologs of Sin1 , Rictor , and Cdk4 are genes required only for abnormal neoplastic glial proliferation but not for glial development . These and other genes within this network may represent important therapeutic targets in human glioma .
Malignant gliomas , neoplasms of glial cells and their precursors , are the most common tumors of the central nervous system ( CNS ) . These tumors typically proliferate rapidly , diffusely infiltrate the brain , and resist standard chemotherapies , properties that render them largely incurable . One key to developing more effective therapies against these tumors is to understand the genetic and molecular logic underlying gliomagenesis . The most frequent genetic lesions in gliomas include mutation or amplification of the Epidermal Growth Factor Receptor ( EGFR ) tyrosine kinase . Glioma-associated EGFR mutant forms show constitutive kinase activity that chronically stimulates Ras signaling to drive cellular proliferation and migration [1] , [2] . Other common genetic lesions include loss of the lipid phosphatase PTEN , which antagonizes the phosphatidylinositol-3 kinase ( PI3K ) signaling pathway , and activating mutations in PIK3CA , which encodes the p110α catalytic subunit of PI3K [1] , [2] . Gliomas often show constitutively active Akt , a major PI3K effector [1] , [2] . However , EGFR-Ras or PI3K mutations alone are not sufficient to transform glial cells , rather multiple mutations that coactivate EGFR-Ras and PI3K-Akt pathways are sufficient to induce glioma [2]–[4] . Understanding the interplay of these mutations and the neurodevelopmental origins of these tumors could lead to new insights into the mechanisms of gliomagenesis . The mammalian brain contains multiple glial cell types that maintain proliferative capacities , including differentiated astrocytes , glial progenitors , and multipotent neural stem cells . EGFR-Ras and PTEN-PI3K signaling regulates many developmental processes in these cell types , particularly proliferation and self-renewal , which are also properties of glioma cells [1] . Although recent hypotheses favor that gliomas arise from multipotent stem cells , data from mouse models demonstrate that differentiated glia , glial progenitors , and stem cells can all produce gliomas in response to genetic lesions found in human gliomas [5] , [6] . Thus , misregulation of these genetic pathways may confer unrestricted proliferative capacities to a range of glial cell types , but how this occurs remains unclear . While many of the same effectors are utilized by EGFR-Ras and PI3K in both glial development and cancer , constitutive activation of these pathways may deploy distinct outputs , not utilized in development , that allow particular cells to escape normal physiological cues that restrain proliferation and self-renewal . The identity of such outputs remains unclear . With these issues in mind , we developed a Drosophila glioma model to facilitate genetic analysis of glial pathogenesis . Drosophila offers many tools for precise manipulation of cell-type-specific gene expression and dissection of multigene interactions . Most human genes , including 70% of known disease genes , have functional Drosophila orthologs [7] . Among the most conserved genes are components of major signal transduction pathways , including many gliomagenic genes . Recently , Drosophila has emerged as a model system for human neurological diseases because the CNS shows remarkable evolutionary conservation in cellular composition and neurodevelopmental mechanisms [8] . Similarly , Drosophila have multiple glial cell types that require the EGFR pathway for their normal development , and these cells appear homologous to mammalian glia in terms of function , development , and gene expression [9] . These similarities between flies and humans make Drosophila an attractive system for modeling gliomas .
Since concurrent activation of EGFR-Ras and PI3K signaling in glial precursors induces glioma in the mouse [4] , we sought to create mutant phenotypes by hyperactivation of these pathways in fly glia and glial precursors . Drosophila has a single functional ortholog each for EGFR ( dEGFR ) , Raf ( dRaf ) , PIK3CA ( dp110 ) , PTEN ( dPTEN ) , and Akt ( dAkt ) , and two functional orthologs for Ras ( dRas85D , dRas64B ) ( www . flybase . org ) . A diagram of the specific mutant forms of dEGFR used in our assays can be found in Figure S1 . We performed glial overexpression assays with the Gal4-UAS system [10] , using the repo-Gal4 driver , which gives sustained UAS-transgene expression in almost all glia , from embryogenesis through adulthood . For glial-specific RNAi , we employed UAS-dsRNA constructs [11] , which we verified with phenotypic tests and/or antibody staining ( see Materials and Methods ) . Glial morphology was visualized with membrane-localized GFP ( CD8GFP ) [12] . Cell number was determined with staining for Repo , a homeobox transcription factor expressed by repo-Gal4 positive glia [13] . Glial-specific coactivation of EGFR-Ras and PI3K stimulated glial neoplasia , giving rise to CNS enlargement and malformation , neurologic defects , and late larval lethality . repo-Gal4-driven co-overexpression of activated dEGFR ( dEGFRλ ) and dp110 ( dp110CAAX ) induced progressive accumulation of ∼50-fold excess glia ( Figure 1A and 1B ) [14] , [15] . dEGFRλ is a constitutively active dEGFR variant in which a lambda dimerization domain replaces the extracellular domain [14] ( Figure S1 ) . Co-overexpression of combinations of dEGFRλ and core components of the PI3K pathway , such as dAkt , induced phenotypes similar to repo>dEGFRλ;dp110CAAX , although phenotypes varied somewhat depending on strength of pathway activation and transgene expression ( Figure S2 and Table S1 ) . Dramatic glial overgrowth also occurred upon co-overexpression of constitutively active dRas ( dRas85DV12 ) or its effector dRaf ( dRafgof ) with dp110CAAX , dAkt , or a dPTENdsRNA , which partially knocked-down dPTEN ( Figure S2 and Table S1 ) . Finally , glial overgrowth in repo>dEGFRλ;dp110CAAX larvae was strongly suppressed by co-overexpression of dPTEN or more moderately by dominant negative dRas85D ( dRas85DN17 ) ( Figure 1D and 1E ) , indicating that Ras activity and excess phospho-inositols are essential for neoplasia . In contrast , glial-specific activation of the EGFR-Ras pathway alone , through overexpression of dEGFRλ or Rafgof , induced 5–10 fold excess glia in the larval brain and later pupal lethality ( Figure 1F and Table S1 ) . dRas85DV12 overexpression induced approximately 5–10-fold excess glia , and these glia were smaller than wild-type or dEGFRλ;dp110CAAX glia . ( Figure 1G ) . dRas85DV12 may be more potent than dEGFRλ because dRas85DV12 can activate endogenous PI3K signaling [16] . Overexpression of dEGFRElp , a classical hypermorphic mutant form of dEGFR [17] , induced excess glial proliferation and neural morphogenesis defects ( Figure S2 ) , but also caused early lethality which precluded examination of dEGFRElp-dp110 interactions . As in mouse models , overexpression of wild-type dEGFR failed to induce excess glia [6] , [17] , and instead retarded CNS growth ( Figure S2 ) . Unlike dEGFRλ , dEGFRWT and dEGFRElp have functional ligand-binding domains ( Figure S1 ) , and may cause additional defects by sequestering ligand otherwise required for normal development [17] . Glial-specific activation of the PI3K pathway alone , either by overexpression of dp110CAAX , dp110wild-type , dAkt , or dPTENdsRNA gave viable animals with relatively normal brains ( Figure 1H , Figure S2 , and Table S1 ) . Therefore , coactivation of the EGFR and PI3K pathways synergize to produce much more severe phenotypes than would be expected if the effects of these pathways were additive . In repo>dEGFRλ;dp110CAAX brains , excess glia emerged in early larval stages and accumulated over 5–7 days . dEGFRλ;dp110CAAX glia severely disrupt the normal cellular architecture of the larval brain ( Figure 1A and 1B and Figure 2A–C ) , lose normal stellate glial morphologies ( Figure 2A–C ) , and generate multilayered aggregations of abnormal glia throughout the brain ( Figure 2A–C ) ; in these ways dEGFRλ;dp110CAAX glia are neoplastic [18] . Like neoplastic epithelial cells , dEGFRλ;dp110CAAX glia ectopically expressed an active form of the matrix metalloprotease dMMP1 ( Figure S3 ) , which can confer an invasive potential [19] , [20] , implying that abnormal dEGFRλ;dp110CAAX glia may be invasive within the brain . Unlike neoplastic epithelia , neoplastic neural cells , such as dEGFRλ;dp110CAAX glia , typically retain expression of genes that regulate neural cell fate , such as Repo [21] , [22] . Relative to controls , many dEGFRλ;dp110CAAX glia showed BrdU incorporation , which marks S-phase cells ( Figure 2D and 2E ) , indicating that neoplastic glia arise from overproliferation . repo>dEGFRλ;dp110CAAX animals also showed reduced BrdU in neuronal precursors ( Figure 2E , data not shown ) , demonstrating that neoplastic glia disrupt neuronal development . The cell cycle is governed by CyclinD-Cdk4 and CyclinE-Cdk2 complexes , which phosphorylate and inactivate Rb proteins , to release E2F activators to stimulate G1-S-phase entry [23] . Kip-type ( p21/p27/p57 ) and Ink-type cyclin-dependent kinase inhibitors antagonize proliferation by inhibiting CyclinE-Cdk2 and CyclinD-Cdk4 , respectively . Cdc25 phosphatases and mitotic cyclins , including CyclinB , promote G2-M progression . Flies have single orthologs each for CyclinE , Cdk2 , CyclinD , Cdk4 , CyclinB , and p21/p27/p57 ( Dap ) , E2F activators ( E2F1 ) and two orthologs for Rb ( Rbf1 and Rbf2 ) and Cdc25 ( Stg and Twe ) but no Ink ortholog [23] . dEGFRλ;dp110CAAX glia showed ectopic expression of dCyclinE and dCyclinB ( Figure 2F–I ) , demonstrating that EGFR and PI3K activity upregulated proteins that promote cell cycle entry and progression . High-grade human gliomas contain highly proliferative anaplastic glia and enlarged pleiomorphic polyploid glia [24] . Similarly , repo>dEGFRλ;dp110CAAX larvae showed accumulation of small , highly proliferative glia that strongly expressed cyclins and labeled with BrdU . repo>dEGFRλ;dp110CAAX larval brains also displayed abnormal polyploid glia , as assessed by DAPI staining ( data not shown ) , and these cells typically expressed only dCyclinE but not dCyclinB , and thereby likely underwent ectopic DNA replication without mitosis ( Figure 2G and 2I , data not shown ) . However , overexpression of dCyclinE-dCdk2 , dCyclinD-dCdk4 , or dE2F1-dDp complexes and/or Rbf1 knock-down did not cause neoplasia , and instead either doubled glial cell numbers or resulted in embryonic lethality ( Figure S4 , data not shown ) . We next examined negative regulators of the cell cycle . dEGFRλ;dp110CAAX glia expressed Rbf1 , but showed little Dap , a result we also observed in wild-type glia ( Figure S5 ) . Dap inhibits dCyclinE-cdk2 complexes [25] , and is transiently expressed in neural progenitors to promote cell cycle exit as they begin differentiation [26] . Dap overexpression completely suppressed repo>dEGFRλ;dp110CAAX glial overgrowth ( Figure 2L ) , demonstrating that glial neoplasia is cell-autonomous and requires dCyclinE-dCdk2 . Similarly , overexpressed Rbf1 and dCyclinE mutations also reduced repo>dEGFRλ;dp110CAAX glial overproliferation ( Figure 2M , data not shown ) . The gross neural morphogenesis defects observed in repo>dEGFRλ;dp110CAAX brains may be secondary to glial overproliferation since these defects were largely prevented by Dap or Rbf1 co-overexpression ( Figure 2J–M ) . In repo>dEGFRλ;dp110CAAX animals , other mutant glia outside of the brain , such as peripheral glia , also became highly proliferative and invasive , and these defects , too , were corrected by Rbf1 or Dap overexpression ( data not shown ) . In controls , Rbf1 or Dap overexpression in wild-type glia inhibited proliferation ( Figure S4 ) , reducing numbers of glia by approximately half . Together , these results suggest that repo-Gal4 glia undergo at least one round of cell division , consistent with published studies [27] , [28] , and this proliferation becomes prolonged by constitutive coactivation of EGFR and PI3K signaling . The phenotype triggered by coactivation of EGFR and PI3K in glia is distinct from other Drosophila brain-overgrowth mutant phenotypes , which involve accretion of excess neurons or neuroblasts [21] , [29] . repo>dEGFRλ;dp110CAAX cells were not transformed into neurons or neuroblasts as they lacked expression of the Elav and Miranda markers ( Figure S6 ) . Lineage-tracing with a Su ( H ) -lacZ neuroblast reporter showed that excess dEGFRλ;dp110CAAX glia did not express LacZ , and thus are not directly derived from larval neuroblasts ( data not shown ) . Moreover , constitutive EGFR-Ras and PI3K signaling does not elicit overgrowth in all neural cell types , as assessed with defined cell-type specific Gal4-drivers ( Table S2 ) . For example , dRas85DV12 overexpression in fly neurons causes defects in fate specification , patterning , and apoptosis [30] , [31] . Co-overexpression of dEGFRλ or dRas85DV12 with dp110CAAX in neurons ( elav-Gal4 , scratch-Gal4 , OK107-Gal4 , and Appl-Gal4 ) and neuroblasts/neuronal precursors ( pros-Gal4 , wor-Gal4 , and 1407-Gal4 ) did not induce overgrowth ( Figure S7 , data not shown ) , even with increased transgene expression from Gal4 amplification [32] . In fact , broad co-overexpression of dEGFRλ and dp110CAAX in neuroblasts and neuronal precursors ( pros-Gal4 ) reduced brain size , perhaps because signaling through these pathways stimulates precocious cell cycle exit of neuronal precursors , as in the developing eye [33] . Furthermore , transient expression of dRas85DV12 or dEGFRλ and dp110CAAX in embryonic glia ( gcm-Gal4 ) also failed to promote glial overgrowth ( Figure S7 ) [27] , demonstrating that sustained activation of these pathways is required for glial overproliferation . Certain glial subtypes , such as oligodendrocyte-like neuropil glia ( Eaat1-Gal4 ) , some astrocyte-like cortex glia ( Nrv2-Gal4 ) , and peripheral perineurial glia ( gli-Gal4 ) also failed to become neoplastic in response to EGFR-Ras and PI3K ( Figure S7 , data not shown ) [34]–[36] . Therefore , neoplastic proliferation is not a uniform cellular response to EGFR and PI3K . Since repo>dEGFRλ;dp110CAAX animals die in 5–7 days , we assessed the proliferative potential of mutant glia using an abdominal transplant assay , a classic test of tumorigencity in flies [37] . Brain fragments from repo>dEGFRλ;dp110CAAX and wild-type larvae were transplanted into young adults . Wild-type transplants grew and survived over 1–6 weeks , but produced few glia ( Figure 3A and 3C ) . dEGFRλ;dp110CAAX mutant glia survived and proliferated into massive tumors that filled the hosts' abdomens , often causing premature death ( Figure 3B ) . Tumors were composed of small glial cells with little cytoplasm ( Figure 3D–F ) . Tumors also contained trachea embedded throughout their mass ( Figure 3D and 3E and Video S1 ) , suggesting that tumors stimulated growth of new trachea or enveloped existing trachea , perhaps in a process akin to tumor angiogenesis . The leading edges of the tumors harbored individual cells invading nearby tissues , such as the ovary ( Figure 3F and Video S2 ) , which is consistent with the ectopic expression of active dMMP1 observed in dEGFRλ;dp110CAAX glia in the larval brain ( Figure S3 ) . However , some tissues , such as the gut , did not contain metastases , implying some degree of selective invasion . Thus , once unconstrained by the larval life cycle , dEGFRλ;dp110CAAX glia fail to exit the cell cycle , continue to proliferate , and form highly invasive tumors , all properties of human cancer cells . To explore the invasive potential of mutant glia , we used FLP-FRT clonal analysis , a technique in which discrete clones of mutant cells are induced in otherwise normal tissues , a situation analogous to somatic tumorigenesis . We used a heat-shock-driven FLP-recombinase to catalyze mitotic recombination between FRT-bearing chromosomes such that a daughter cell ( and all of its clonal progeny ) initiated expression of GFP and UAS-containing transgenes only in repo-Gal4-expressing cells [12] . Clones were induced late in development , from mitotic founder cells , and were examined in adults . We could not definitively determine if clones were derived from single cell events since our study of these clones was retrospective , but given the frequency of control clone induction , many mutant clones likely originated from single cells . In wild-type controls , we observed clones in 68% of brains examined ( N = 149 ) . Of the brains with clones , 75% had 1–3 clones , and 83% of these clones consisted of 1–3 cells of the same glial subtype ( Figure 4A and 4B ) . Glial clones overexpressing dRas85DV12 , dEGFRλ , or dEGFRElp alone typically contained 2-fold more cells than wild-type ( dRas85DV12 shown in Figure 4C ) . To examine PI3K signaling , we used a dPTEN null allele , which became homozygous in FLP-FRT clones . dPTEN−/− glia did not overgrow , but did show aberrant cytoplasmic projections ( Figure 4D ) , perhaps reflecting dPTEN function in the cytoskeleton [38] . To coactivate EGFR-Ras and PI3K in glial clones , dRas85DV12 , dEGFRElp , or dEGFRλ was overexpressed within dPTEN−/− glia , using repo-Gal4 . We observed overgrown and invasive dRas85DV12;dPTEN−/− and dEGFRElp;dPTEN−/− clones ( Figure 4E–G ) . dEGFR;dPTEN−/− clones were less affected ( Figure 4F and 4G ) , consistent with the larval overexpression of dRas85DV12 giving more severe growth phenotypes . Tumor-like overgrowth was only observed in dEGFR-dRas85D;dPTEN−/− cells , illustrating that chronic EGFR-Ras activation and PTEN loss can cause cell-autonomous over-proliferation . Cells from these double mutant clones appeared to invade the brain , typically following fiber tracts , and sometimes induced the formation of trachea ( Figure 4E ) . Tumor-like growths of dEGFR-dRas85D;dPTEN−/− cells often penetrated deep into the brain , as exemplified by Videos S3 and S4 which show an animated 88 µm thick confocal z-stack of the dRas85DV12;dPTEN−/− clone in Figure 4E compared to a 16 µm thick z-stack of a wild-type control clone in Figure 4A . These phenotypes were reminiscent of invasion and angiogenesis in human gliomas [24] . We more commonly observed smaller dEGFR-dRas85D;dPTEN−/− clones composed of relatively differentiated , enlarged glia with diffusely invasive projections ( Figure 4G ) ; these clones likely derive from glia that differentiated prior to achieving sufficient EGFR-Ras transgene expression , and are consistent with findings that not all glial subtypes become neoplastic . Since neoplastic larval glia were concentrated in the outer anterior central brain and developing optic lobe , they may be derived from glial progenitor cells present in these regions [27] , [28] , [39] . To create clones from a discrete subpopulation of glial progenitors , we used an eyeless ( ey ) -promoter driven FLP-recombinase ( ey-FLP ) , which is active in ey-expressing glial progenitors in the optic lobe [27] . Single mutant ey-FLP clones of dPTEN−/− , dRas85DV12 , or dEGFRElp cells contained a modest number of excess and abnormal glia relative to wild-type controls ( Figure 4H–J and Figure S8 ) . In contrast , the double mutant dRas85DV12;dPTEN−/− , dEGFRElp;dPTEN−/− , or dEGFRλ;dPTEN−/− ey-FLP clones , which emerge as approximately tens of cells in 3rd instar larval brains ( Figure 4K ) , became large invasive tumors composed of hundreds to thousands of cells in adults ( Figure 4L and Figure S8 ) . To address the function of EGFR-Ras and PI3K in glioma , we analyzed the genetic basis of glial pathogenesis in our repo>dEGFRλ;dp110CAAX model , as this model shows robust neoplasia , similarity to human tumor genotypes , and sensitivity to dEGFR and dPTEN gene dosage ( data not shown ) . EGFR-Ras signaling can promote proliferation through Erk kinase-mediated induction of nuclear targets . In repo>dEGFRλ;dp110CAAX brains , mutant glia showed high levels of nuclear , activated di-phospho-Erk relative to wild-type glia in control brains ( Figure 5A and 5B ) . In flies , Erk activity can induce expression of PntP1 , an ETS-family transcription factor encoded by the pointed ( pnt ) locus [40] , [41] . PntP1 , which is expressed in embryonic glia and is required for their normal development [40] , was upregulated in dEGFRλ;dp110CAAX glia ( Figure 5C and 5D ) . High levels of PntP1 can be detected normally in neuronal progenitors ( data not shown ) , suggesting that it promotes a proliferative progenitor state . In developing eye tissue , EGFR-Ras-Erk signaling induces Pnt proteins to stimulate G2-M cell cycle progression through direct upregulation of Stg ( Cdc25 ortholog ) expression [41] . Glial-specific RNAi knock down of pnt reduced Stg expression and completely suppressed dEGFRλ;dp110CAAX neoplasia ( Figure 5E–H ) , demonstrating that Pnt proteins are required for both Stg expression and neoplastic overproliferation in dEGFRλ;dp110CAAX glia . Notably , in repo>pntdsRNA;dEGFRλ;dp110CAAX brains , glia maintained their fate , as evidenced by repo-Gal4 and Repo expression . Stg itself was rate limiting for glial neoplasia . Reduction of stg with a mutation or a stgdsRNA partially suppressed the repo>dEGFRλ;dp110CAAX phenotype , whereas overexpressed Stg synergistically enhanced neoplasia ( Figure 5E and Figure S9 , data not shown ) . In contrast , Stg overexpression alone increased glial cell numbers approximately 2-fold , and could not induce neoplasia when combined with PI3K effectors ( data not shown ) . Thus , dEGFRλ;dp110CAAX induces neoplasia via coordinated stimulation of G1-S entry through dCyclinE , and G2-M progression through Stg , both of which are EGFR-Ras dependent outputs [41] , [42] . We sought to determine which PI3K effectors contribute to the repo>dEGFRλ;dp110CAAX phenotype . Genetic reduction of dAkt , a major target of PI3K signaling , with a dAktdsRNA or a mutant allele strongly suppressed repo>dEGFRλ;dp110CAAX glial neoplasia ( Figure 6A–C , data not shown ) . Therefore , Akt is necessary for the outcome of EGFR-PI3K coactivation . Many Akt effectors are implicated in glioma , and we tested orthologs of these loci in our model ( Table S3 ) . Tor , a kinase that promotes cell growth and proliferation , is a key Akt target . In glioma models , coactivation of EGFR-Ras and PI3K stimulates Tor , and in humans , Tor activity is correlated with poor patient prognosis [43] , [44] . We tested the single Drosophila Tor ortholog , dTor , by genetically reducing dTor activity in repo>dEGFRλ;dp110CAAX larvae with a viable combination of hypomorphic dTor alleles or by co-overexpression of dominant negative dTor 45 , 46 . Both of these manipulations reduced glial overgrowth ( Figure 6D , data not shown ) . In flies and mammals , Tor exists in two different signaling complexes , TORC1 and TORC2 . TORC2 , a complex including Tor and the Sin1 and Rictor regulatory proteins , directly phosphorylates Akt , creating a positive feedback loop that fully activates Akt [47] . In mouse , Sin1 and Rictor mutants die early due to extraembryonic defects , but dSin1 and dRictor mutant flies are viable as homozygous nulls , allowing us to remove TORC2 function genetically [47] , [48] . dSin1−/−; repo>dEGFRλ;dp110CAAX larval brains showed a near-wild type phenotype ( Figure 6E ) . Results were similar with a dRictor null allele and a dSin1dsRNA ( data not shown ) . dSin1−/− and dRictor−/− mutants display reduced Akt-dependent phosphorylation and inactivation of dFoxO [49] , the single Drosophila ortholog of FoxO transcription factors . This suggests that TORC2 loss might antagonize glial neoplasia through dFoxO upregulation . However , excess dFoxOSA , which is resistant to dAkt phosphorylation [45] , only partially suppressed repo>dEGFRλ;dp110CAAX glial overproliferation ( Figure 6F ) , arguing that TORC2 has additional roles . Notably , on their own , dSin1−/− and dRictor−/− mutant flies did not show any detectable glial defects ( Figure 6G , data not shown ) . Thus , TORC2 is dispensible for normal glial development , but is necessary for dEGFRλ;dp110CAAX glial neoplasia . TORC1 , a complex including Tor and the Raptor regulatory protein , drives cellular growth by stimulating protein synthesis through its effectors S6 kinase and the eIF-4E translation initiation factor [47] . Akt and Erk stimulate TORC1 through phosphorylation and inactivation of the TSC1-TSC2 protein complex , which activates Rheb , and stimulates TORC1 kinase activity . We tested TORC1 function by glial-specific overexpression of dsRNAs for the single Drosophila orthologs of Raptor ( dRaptor ) , S6-kinase ( dS6K ) , and eIF4E ( deIF4E ) ; these all significantly reduced accumulation of dEGFRλ;dp110CAAX mutant glial cells , but only caused mild glial hypoplasia in controls ( Figure 6I–K and Figure S10 ) . Co-overexpression of d4EBP , a deIF4E antagonist and dFoxO target gene , also blocked glial neoplasia ( Figure 6L ) . Glial-specific RNAi of dTSC1 enhanced repo>dEGFRλ;dp110CAAX glial overgrowth ( Figure 6H ) . However , overexpression of dTSC1dsRNA , dRheb , activated dS6K ( dS6Kact ) , or deIF4E alone did not produce glial overproliferation , even though these constructs can mimic TORC1 activation ( Figure S10 , data not shown ) [47] , [49] . Thus , TORC1 activity is necessary for EGFR-PI3K-driven glial neoplasia , but is not sufficient . Moreover , neither deIF4E nor dS6Kact produced glial neoplasia when co-overexpressed with dEGFRλ ( data not shown ) , illustrating that additional dTor-dependent outputs synergize with dEGFR signaling to drive neoplasia . dTor coordinates increased translation , mediated by dS6K and deIF4E , with expression of cell cycle regulators and ribosomal components , through dMyc , the single Drosophila ortholog of the Myc bHLH transcription factors [49] . Within developing epithelial tissues , dMyc is required for TORC1-dependent growth and can substitute for dTor activity [49] , [50] . Myc protein levels can also be posttranslationally upregulated by EGFR-Ras-Raf signaling [16] , [51] . Thus , we suspected that dMyc might mediate signal integration between EGFR-Ras and PI3K . dEGFRλ;dp110CAAX glia showed high levels of nuclear dMyc compared to wild-type glia ( Figure 7A and 7B ) . dMyc was also highly expressed in wild-type neuroblasts ( Figure 7A ) , suggesting that dMyc promotes a proliferative progenitor state [21] . Genetic reduction of dMyc with a dsRNA or a single loss-of-function allele strongly suppressed dEGFRλ;dp110CAAX glial neoplasia ( Figure 7D , data not shown ) . In fact , some dMyc+/−; repo>dEGFRλ;dp110CAAX animals were rescued to viability , indicating that dMyc is an essential rate-limiting output of EGFR-PI3K coactivation . Myc proteins activate transcription through heterodimerization with the bHLH Max . Max activity was also required for glial neoplasia; a dsRNA for dMax , the single Drosophila Max ortholog , strongly suppressed repo>dEGFRλ;dp110CAAX overgrowth ( Figure 7E ) . dMyc-dMax heterodimers promote proliferation by activating expression of multiple cell cycle genes , including dCyclinD and dCdk4 [52] . dCyclinD expression , which is high in repo>dEGFRλ;dp110CAAX glia , was inhibited by a dMycdsRNA ( Figure 7F–H ) , suggesting that dMyc reduction suppresses glial neoplasia through reduced dCyclinD-dCdk4 activity . To test this we used loss-of-function mutations in dCdk4 , which are viable [23] . dCdk4−/−; repo>dEGFRλ;dp110CAAX larvae showed near complete absence of excess glia and rescue of glial morphogenesis defects ( Figure 7I ) . Moreover , a dCdk4dsRNA and dCyclinDdsRNA suppressed repo>dEGFRλ;dp110CAAX ( data not shown ) . Thus , dCyclinD-dCdk4 is essential for EGFR-PI3K glial neoplasia , although dCdk4 is not required for development ( Figure 7J ) . dMyc is necessary for glial neoplasia , but is not sufficient when overexpressed alone ( Figure 8C ) . dMyc-overexpressing glia showed polyploidy ( Figure 8C , data not shown ) , indicating that these cells undergo DNA replication without mitosis , but require additional signals for cell cycle progression . In contrast , co-overexpression of dMyc with dEGFRλ produced a phenotype on par with that of repo>dEGFRλ;dp110CAAX ( Figure 8D ) , indicating that dMyc overexpression can substitute for PI3K activation and promote neoplasia when combined with EGFR signaling . Given that the dMyc targets dCyclinD-dCdk4 are required for dEGFRλ;dp110CAAX neoplasia , we tested whether dCyclinD-dCdk4 overexpression could cooperate with dEGFRλ . Additionally , we tested loss of Rbf1 , the only known dCdk4 substrate that controls proliferation [23] . repo>dCyclinD;dCdk4;dEGFRλ and repo>Rbf1dsRNA;dEGFRλ animals showed glial overgrowth , but did not accumulate as many cells as repo>dEGFRλ;dp110CAAX animals ( Figure 8E and 8F ) . Thus , glia likely require additional dp110 or dMyc effectors to undergo full neoplastic proliferation . Other known PI3K and dMyc-dMax target genes that promote proliferation include ribosomal proteins and translation regulators [49] , [52] , such as eIF4E , which is highly expressed in dEGFRλ;dp110CAAX glia and required for neoplasia ( Figure 6 and Figure S11 ) . Our data imply that PI3K , dMyc , and dCyclinD-dCdk4 exist in a linear system , in which Rbf1 inactivation by dCyclinD-dCdk4 is one direct output of dp110CAAX or dMyc . However , in high-grade glioma , Rb loss co-occurs with EGFR and PTEN mutations [2] , implying that these mutations cooperate to promote gliomagenesis . To explore interactions between Rb and EGFR-PI3K , we created the triple mutant repo>Rbf1dsRNA;dEGFRλ;dp110CAAX . These animals displayed exacerbated glial neoplasia , with a substantial increase in small anaplastic-like glia throughout the CNS ( Figure 8G ) . This synergistic interaction likely derives from derepression of dE2F1 upon Rbf1 loss , and concomitant increased expression of dE2F1 target genes , including Stg and dCyclinE [53] . Increased dCyclinE and Stg expression may accelerate cell cycle progression , perhaps through increased dCdk2 and dCdk1 activity and/or truncated G1 and G2 gap phases caused by constant dCyclinE and Stg protein levels [23] . Consistent with this , we observed increased dCyclinE expression in Rbf1dsRNA;dEGFRλ;dp110CAAX glia relative to dEGFRλ;dp110CAAX glia ( Figure 8H and 8I ) , and co-overexpression of Stg or dCyclinE-dCdk2 with dEGFRλ;dp110CAAX synergistically exacerbated glial neoplasia , yielding phenotypes similar to repo>RbfdsRNA;dEGFRλ;dp110CAAX ( Figure 5E and Figure S9 , data not shown ) . To assess dCdk2 activity in repo>Rbf1dsRNA;dEGFRλ;dp110CAAX brains compared to repo>dEGFRλ;dp110CAAX brains , we stained for phospho-MPM2 ( Figure S12 ) , which detects nuclear foci in cells with active dCyclinE-dCdk2 complexes [54] . Phospho-MPM2 foci were present in glia of both genotypes , although repo>Rbf1dsRNA;dEGFRλ;dp110CAAX brains appeared to have a higher density of glia with phospho-MPM2 foci ( Figure S12 ) , suggesting that expanded expression of dCyclinE results in broader activation of dCdk2 . Thus , while PI3K and Rbf1 act in a common genetic pathway linked by dCyclinD-dCdk4 , Rbf1 loss nevertheless synergizes with mitogenic stimulation from combined EGFR and PI3K signaling , and this synergy emerges from increased expression of dCyclinE and Stg , rate-limiting regulators of the cell cycle .
We show that constitutive coactivation of EGFR-Ras and PI3K signaling in Drosophila glia and glial precursors gives rise to neoplastic , invasive cells that create transplantable tumor-like growths , mimicking human glioma , and mirroring mouse glioma models . This represents a robust organotypic and cell-type specific Drosophila cancer model in which malignant cells are created by mutations in the signature genes and pathways thought to be driving forces in a homologous human cancer . This was not necessarily an expected result since fly and human glia show many biological differences despite displaying important similarities [9] , [55] , [56] . Through genetic analysis of our model , we identified crucial downstream effectors of EGFR and PI3K signaling , many of which are mutated and/or activated in human glioma . These effectors act in a combinatorial network to coordinately stimulate cell cycle entry and progression , block cell cycle exit , and promote inappropriate cellular growth and migration ( Figure 8J ) . Pathways within this network , while interdependent , act synergistically , rather than additively . Thus , Drosophila shows evolutionary conservation of oncogene cooperation . At least four pathway circuits are necessary for glial neoplasia initiated by EGFR and PI3K signaling , including dRas and dMyc circuits , which induce dCyclinE and dCyclinD to drive cell cycle entry , a Pnt circuit , which induces Stg to promote cell cycle progression , and a Tor-eIF4E-S6K pathway , which provides protein translation necessary for proliferation and growth ( Figure 8J ) . When activated individually , these pathways fail to elicit glial neoplasia , implying a requirement for coordinated stimulation of multiple effectors and inactivation of negative regulators . Orthologs for many of the genes within these pathways , such as dRictor , are implicated in human glioma , although specific roles for some , such as ETS transcription factors , have not been defined despite their expression in glioma [2] , [57] , [58] . While many of these genes are known EGFR and PI3K pathway components , we did not necessarily expect them to be required for EGFR and PI3K dependent glial neoplasia . Indeed , we have tested many other pathway components and outputs , such as Jun kinase , that did not significantly suppress repo>dEGFRλ;dp110CAAX phenotypes upon reduced function ( unpublished data ) . Coactivation of EGFR and PI3K signaling upregulates dMyc , which is necessary for glial neoplasia . This is consistent with findings that , in flies and mammals , EGFR-Ras , PI3K , and Tor signaling upregulate Myc protein levels [16] , [49] , [51] , [59] , [60] . Myc oncogenes are well-known to cooperate with RTK-Ras signaling to drive neoplastic transformation [51] , and we demonstrate that this property of Myc is conserved in flies . We also observed sensitivity to reduced Myc gene dosage in our glioma model , which has also been recently documented in a mouse model of PTEN-dependent glioma [61] . c-myc is commonly amplified in gliomas [62] , implying that Myc is rate limiting , and c-myc amplification may be selected for this reason . D-cyclins , established Myc target genes , and Cdk4 are also commonly amplified and/or overexpressed in gliomas [1] , [51] . We observed dMyc-dependent dCyclinD overexpression , and a requirement for dCyclinD-dCdk4 in repo>dEGFRλ;dp110CAAX neoplasia , although dCdk4 itself is not required for normal glial proliferation . Together with our analysis of TORC2 , this illustrates that oncogenic EGFR-PI3K co-opts effectors that do not control normal glial development . Similarly , cdk4−/− mutant mice show normal proliferation in many tissues , but are resistant to ErbB-2-driven breast cancers [63] , [64] . Our data argue that Cdk4 activity is a key tumor-specific rate-limiting output of EGFR and PI3K signaling in glioma as well . In contrast to glia , coactivation of EGFR-Ras and PI3K in neuroblasts , which are fly neural stem cells , does not promote unchecked proliferation , despite the fact that neuroblasts express dMyc and are capable of undergoing neoplastic transformation in response to other genetic mutations [21] . Thus , in Drosophila , neither a neural stem cell fate nor Myc activity confer competence to undergo EGFR-PI3K neoplastic transformation . Rather , our results suggest that neoplastic cells arise from committed glial progenitors: dEGFR-dRas85D;dPTEN−/− clones derived from progenitor cells produce large tumors , and anaplastic cells in repo>dEGFRλ;dp110CAAX brains are concentrated in regions enriched for glial progenitors . Notably , regulated developmental signaling through the EGFR pathway promotes proliferation of normal Repo-expressing glial progenitors [27] , and our results show that constitutive EGFR and PI3K signaling prolongs this proliferative progenitor state . Further studies of Drosophila glial progenitors and glioma-like cells may illuminate the cellular origins of human gliomas , which are thought to arise from progenitor-like glial cells . Moreover , our results argue that cell-type specific factors govern glial neoplasia . One such factor may be Dap , the single p21/p27 ortholog , which is normally expressed in only 5% of all glia ( Figure S4 ) . Perhaps glial progenitors do not express Dap , whereas neuronal progenitors do [26] , and this underlies susceptibility to transformation by EGFR-Ras and PI3K . Dap is highly regulated in a cell-type specific manner [26] , and studies of Dap regulation in glia may further illuminate the genetic origins of glioma , especially given that lack of p21 expression may underlie the tumorigenic response of mammalian glial progenitors to constitutively active EGFR [65] . While EGFR-Ras and PI3K are commonly upregulated in gliomas and experimental models demonstrate that these pathways are required for tumorigenesis , therapies that target EGFR and PI3K signaling have proven disappointing . This discrepancy between clinical and experimental data has many possible explanations . For example , recent studies have demonstrated that EGFR inhibitors are attenuated by particular mutations found in glioma cells , such as PTEN loss or RTK co-amplification [2] . Addressing these and other possibilities remains a challenge that dictates a need for new experimental models . The results presented here establish Drosophila as a viable model system for the study of glioma , offering a complex organismal system for rapidly identifying and evaluating therapeutic targets using genetic approaches . Such a system may be especially useful for distinguishing those genetic mutations and pathways that drive tumorigenesis from the large number of genes that show mutations and altered expression in glioblastomas uncovered by recent genomic analyses of patient samples [66] , [67] . Our studies have already identified key rate-limiting genes , such as dCyclinE , Stg , and dMyc , and genes only required for abnormal neoplastic glial proliferation , such as dSin1 , dRictor , and dCdk4 , which may represent important therapeutic targets in human gliomas .
Flies were cultured at 25°C . All genotypes were established by standard genetics . To assess larval brain overgrowth phenotypes , embryos were collected for 6–24 hrs , grown for 120–140 hrs , and wandering 3rd instar larvae were selected for dissection . Stocks were obtained from the Bloomington Stock Center unless otherwise noted . Other than UAS-PTENdsRNA lines from Bloomington , all UAS-dsRNA lines were obtained from the VDRC stock center [11] . The following stocks were obtained from other investigators: UAS-dEGFRλ ( T . Schubach ) , UAS-dEGFRElp , UAS-dEGFRwild-type ( N . Baker ) , UAS-dPTEN , FRT40A dPTEN2L117 , UAS-dFoxOSA ( S . Oldham ) , UAS-dap ( I . Hariharan ) , UAS-dp110wild-type , UAS-dCycD , UAS-dCdk4 , UAS-dMyc , dMyc4 ( B . Edgar ) , UAS-Rbf1 ( N . Dyson ) , appl-Gal4 ( K . Finley ) , pros-Gal4 ( B . Ohlstein ) , wor-Gal4 ( C . Doe ) , gcm-Gal4 ( V . Hartenstein ) , dTor2L7 , dTorl ( 2 ) k17004 ( R . Bodmer ) , dRictorΔ2 ( S . Cohen ) , and stgCB03726 ( A . Spradling ) UAS-dMycdsRNA , UAS-TSC1dsRNA lines were validated in prior publications [48] . UAS-dsRNA lines were crossed to actin-Gal4 , ey-Gal4 , and GMR-Gal4 to assess phenotypes . Lines that showed phenotypes inconsistent with known phenotypes for their target genes were excluded from analysis . Gene knock-down in repo-Gal4 glia was verified with immunohistochemical stains for the following constructs: UAS-dMycdsRNA , UAS-dAktdsRNA , UAS-dS6KdsRNA , UAS-eIF4EdsRNA , UAS-Rbf1dsRNA , and UAS-pntdsRNA ( Figure S11 ) . Larval brains were dissected into sterile PBS , washed , and cut into fragments . Abdominal incisions were made in virgin female hosts and single brain fragments were inserted . Hosts were cultured for 1–6 weeks , dissected and fixed in 4% paraformaldehyde , incubated in 10% sucrose and embedded in O . C . T . Thick 50 µm sections were stained as described below . For hs-FLP clones , genotypes are indicated in figure legends . Flies were initially grown at 18°C or 20°C to minimize spontaneous clones , which occurred at a low frequency during late larval-pupal stages . 3rd instar larvae , 0–48 hr pupae , or 0–2 day old young adults were treated with heat shock to induce clones and subsequently cultured at 25°C for 1–4 weeks . For ey-FLP clones , flies were cultured at 25°C continuously . Larval tissue was fixed for 30–50 minutes in 1×PBS 4% paraformaldehyde . Adult brains were fixed for 1–2 hr in 1×PBS 4% paraformaldehyde or in PLP with 2% paraformaldehyde . For BrdU labeling , larvae were cultured in food with 1 mg/ml BrdU for 4–6 hrs , and fixed larval brains were treated with 2 N HCl for 30 minutes followed by DNase for 1 hr . Stains were performed in 1×PBS 10% BSA with 0 . 3% Triton-X100 for larval brains and 0 . 5% Triton-X100 for adult samples . The following antibodies were obtained from the Developmental Studies Hybridoma Bank and diluted 1∶5–1∶10: 8D12 anti-Repo , anti-dMMP1 , anti-dCyclinB , anti-Elav , and 40-1a anti-lacZ . Larval and/or adult brains were also stained with rabbit anti-Repo ( G . Technau , 1∶500 ) , rat anti-dCyclinE ( H . Richardson , 1∶100 ) , anti-BrdU ( BD , 1∶100 ) , rat anti-Miranda ( C . Doe , 1∶100 ) , mouse anti-diphospho-Erk ( Sigma , 1∶200 ) , mouse anti-Rbf1 ( N . Dyson , 1∶5 ) , mouse anti-Dap ( I . Hariharan , 1∶10 ) , rabbit anti-PntP1 ( J . Skeath , 1∶500 ) , rabbit anti-eIF4E ( P . Lasko , 1∶100 ) , rabbit anti-dMyc ( D . Stein , 1∶1000 ) , and anti-phospho-MPM2 ( Upstate Biotechnology , 1∶200 ) . Anti-HRP-Cy5 and anti-HRP-Cy3 ( Jackson Labs ) were used at 1∶250–1∶500 . Secondary antibodies were conjugated to Cy3 ( Jackson Labs ) or Alexa-488 or Alexa-647 ( Molecular Probes ) . Actin was visualized with Rhodamine-labeled phalloidin ( Invitrogen ) . Brains were imaged as whole mounts on a Zeiss LSM 510 confocal system . Images were analyzed in Zeiss LSM Image Browser and processed in Photoshop CS3 . For experiments in which protein levels were compared between genotypes , all sample preparation , histochemistry , imaging , and image processing was performed in parallel in the same manner . | Malignant gliomas , tumors composed of glial cells and their precursors , are the most common and deadly human brain tumors . These tumors infiltrate the brain and proliferate rapidly , properties that render them largely incurable even with current therapies . Mutations in genes within the EGFR-Ras and PI3K signaling pathways are common in malignant gliomas , although how these genes specifically control glial pathogenesis is unclear . To investigate the genetic basis of this disease , we developed a glioma model in the fruit fly , Drosophila melanogaster . We found that constitutive coactivation of the EGFR-Ras and PI3K pathways in Drosophila glia gives rise to highly proliferative and invasive neoplastic cells that create transplantable tumor-like growths , mimicking human glioma . This represents a robust cell-type-specific Drosophila cancer model in which malignant cells are created by mutations in genetic pathways thought to be driving forces in a homologous human cancer . Genetic analyses demonstrated that EGFR-Ras and PI3K induce fly glial neoplasia through activation of a combinatorial genetic network composed , in part , of other genetic pathways also commonly mutated in human glioma . This network acts synergistically to coordinately stimulate cellular proliferation , protein translation , and inappropriate migration . Rate-limiting genes within this network may represent important therapeutic targets in human glioma . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology/neuronal",
"and",
"glial",
"cell",
"biology",
"cell",
"biology/cell",
"signaling",
"oncology/neuro-oncology",
"cell",
"biology/cell",
"growth",
"and",
"division",
"genetics",
"and",
"genomics/disease",
"models",
"neuroscience/neurodevelopment",
"developmental... | 2009 | A Drosophila Model for EGFR-Ras and PI3K-Dependent Human Glioma |
Microtubule-based kinesin motors have many cellular functions , including the transport of a variety of cargos . However , unconventional roles have recently emerged , and kinesins have also been reported to act as scaffolding proteins and signaling molecules . In this work , we further extend the notion of unconventional functions for kinesin motor proteins , and we propose that Kif13b kinesin acts as a signaling molecule regulating peripheral nervous system ( PNS ) and central nervous system ( CNS ) myelination . In this process , positive and negative signals must be tightly coordinated in time and space to orchestrate myelin biogenesis . Here , we report that in Schwann cells Kif13b positively regulates myelination by promoting p38γ mitogen-activated protein kinase ( MAPK ) -mediated phosphorylation and ubiquitination of Discs large 1 ( Dlg1 ) , a known brake on myelination , which downregulates the phosphatidylinositol 3-kinase ( PI3K ) /v-AKT murine thymoma viral oncogene homolog ( AKT ) pathway . Interestingly , Kif13b also negatively regulates Dlg1 stability in oligodendrocytes , in which Dlg1 , in contrast to Schwann cells , enhances AKT activation and promotes myelination . Thus , our data indicate that Kif13b is a negative regulator of CNS myelination . In summary , we propose a novel function for the Kif13b kinesin in glial cells as a key component of the PI3K/AKT signaling pathway , which controls myelination in both PNS and CNS .
Myelination is a multistep process that includes axon recognition and contact , ensheathment , and myelin biogenesis . In this process , discrete sets of proteins and lipids are specifically assembled to generate and maintain distinct structural and functional domains necessary for nerve function [1–5] . During myelination , positive and negative regulators must be tightly controlled so that myelin thickness is strictly proportional to axonal diameters . However , the molecular mechanisms that promote and regulate myelination as well as the molecular machineries responsible for the transport and targeting of vesicles during myelin biogenesis are largely unknown . For example , Kif1b is the only motor protein identified thus far implicated in central nervous system ( CNS ) myelination in Danio rerio ( zebrafish ) [6] . We previously reported that in Schwann cells the Kif13b motor protein ( also known as guanylate kinase-associated kinesin [GAKIN] in humans ) is part of a complex that titrates membrane formation during Schwann cell myelination [7] . We found that Kif13b interacts with the Discs large 1 ( Dlg1 ) scaffold in Schwann cells and that the downregulation of either Kif13b or Dlg1 expression in Schwann cell/dorsal root ganglia ( DRG ) neuron co-cultures decreases myelination in vitro [7] . Another study independently reported that Dlg1-silenced Schwann cells in vitro showed migration defects and reduced expression of the polarity protein Par3 [8] . Occasionally , silenced cells overcame their migration defect and myelinated , but the resulting myelin segments were thicker than those of controls , which indicated Dlg1 as a negative regulator of myelin sheath thickness [8] . This role was further assessed in vivo , as we and others subsequently reported that mouse nerves lacking Dlg1 expression specifically in Schwann cells have hypermyelination , myelin outfoldings , and demyelination as a consequence of myelin instability [8 , 9] . Dlg1 is thought to act in complex with phosphatase and tensin homolog ( PTEN ) to reduce AKT ( v-AKT murine thymoma viral oncogene homolog ) activation; thus , it is a brake on myelination [8] . Kif13b kinesin is a plus end motor protein that mediates the transport of several cargos in polarized cells [10–16] . In PC12 cells , Kif13b negatively regulates centaurin-α1/PIP3BP ( phosphatidylinositol-3 , 4 , 5-trisphosphate binding protein ) , a GTPase activating protein ( GAP ) for Arf6 ( ADP-ribosylation factor ) GTPase and promotes Arf6 plasma membrane activation [16] . In neurons , Kif13b transports centaurin-α1/PIP3BP and PIP3 to the tip of neurites to promote neuronal polarity [11] . To further investigate the function of the Kif13b/Dlg1 complex in myelination in vivo , we generated a novel Kif13b floxed allele and conditional knock-out mouse models with specific ablation of Kif13b or Dlg1 in either Schwann cells or oligodendrocytes . Here , we report that Kif13b has opposite roles in the control of myelination in the peripheral nervous system ( PNS ) and CNS . Our data indicate that in Schwann cells , Kif13b interacts with p38γ mitogen-activated protein kinase ( MAPK ) to promote phosphorylation and ubiquitination of Dlg1 . Consistent with this observation , loss of Kif13b results in reduced levels of p38γ MAPK , increased Dlg1 expression , and reduced myelin thickness . Finally , we report that Kif13b also controls Dlg1 function in oligodendrocytes by promoting its negative regulation . However , our data indicate that , in contrast to Schwann cells , Dlg1 does not reduce but rather enhances AKT activation in oligodendrocytes . Thus , Kif13b is a novel negative regulator of CNS myelination .
We previously reported that , in the peripheral nerve , Kif13b is mainly detected in cytosolic compartments of myelin-forming and non-myelin-forming Schwann cells [7] . To investigate the role of Kif13b in Schwann cells in vivo , we generated a Kif13bFloxed ( hereafter , Kif13bFl ) allele in which exon 6 was flanked by lox-P sites . Using the Cre/loxP technology , excision of exon 6 produces a frameshift leading to the introduction of a premature stop codon and to nonsense-mediated mRNA decay ( Fig 1A–1C ) . To ablate Kif13b specifically in Schwann cells , we generated Kif13bFl/Fl P0-Cre mice , in which the myelin protein zero ( MPZ ) promoter drives Cre recombinase expression specifically in Schwann cells , starting from E13 . 5 [17 , 18] . Deletion of exon 6 was documented by PCR analysis on DNA from the sciatic nerve , where a recombination band of 378 bp was specifically detected ( Fig 1D ) . Kif13b protein expression was ablated in sciatic nerve lysates from Kif13bFl/Fl P0-Cre mice , thus also confirming that Kif13b is mainly expressed by Schwann cells in the nerve ( Fig 1E ) . We then analyzed Kif13bFl/Fl P0-Cre sciatic nerves starting at P30 by performing semithin section and ultrastructural analyses . In mutant nerves , we noted a higher number of fibers displaying Schwann cell nuclei and the surrounding cytoplasm , suggesting a shorter internodal length ( Fig 2A ) . Consistent with this , we found that Kif13bFl/Fl P0-Cre quadriceps nerves had indeed a higher percentage of fibers with shorter internodes , particularly in the range between 500 and 600 μm ( Fig 2B ) . Cajal bands are cytoplasmic channels located at the abaxonal surface of myelinated fibers and are involved in the biosynthesis and assembly of myelin [3] . Ablation of the Schwann cell protein Periaxin disrupts Cajal bands and is also associated with reduced longitudinal growth of Schwann cells [19] . However , subsequent work from the same group has shown that loss of Cajal bands as a result of Drp2 ablation causes focal hypermyelination and concomitant demyelination [20] . We analyzed Kif13bFl/Fl P0-Cre quadriceps nerves , but we did not detect major differences in Cajal band structures between mutant and control nerves ( Fig 2C ) . Our findings are consistent with the view that the longitudinal growth of Schwann cells does not correlate with Cajal band integrity [20] . As myelin thickness is proportional to axonal diameter and internodal length [3] , we evaluated myelin thickness in Kif13bFl/Fl P0-Cre nerves by performing ultrastructural analysis . By measuring the g-ratio—the ratio between axonal diameter and fiber diameter—we observed reduced myelin thickness in mutant quadriceps nerves at P30 , which displayed increased g-ratio values as compared to controls ( ultrastructural analysis , Kif13bFl/Fl P0-Cre , 0 . 75 ± 0 . 008 , 575 fibers; Kif13bFl/+ , 0 . 70 ± 0 . 009 , 516 fibers , n = 3 animals per genotype , p = 0 . 03 ) . At P20 , myelin thickness was normal in Kif13bFl/Fl P0-Cre nerves , suggesting that myelination is not delayed in this mutant ( ultrastructural analysis , g-ratio values: Kif13bFl/Fl P0-Cre , 0 . 715 ± 0 . 007 , 400 fibers; Kif13bFl/+ , 0 . 71 ± 0 . 004 , 403 fibers , n = 3 animals per genotype , p = 0 . 49 ) . Reduced myelin thickness was still present in nerves of older mice at 8 mo , as g-ratio values were increased in mutant nerves ( Fig 2D ) . Finally , following crush nerve injury , remyelinating Kif13bFl/Fl P0-Cre nerves also displayed thinner myelin ( S1 Fig ) . In conclusion , our data indicate that loss of Kif13b specifically in Schwann cells affects longitudinal and radial myelin growth during development and remyelination after injury . Of note , myelination is not delayed in Kif13b mutant nerves at early stages of development , suggesting that Kif13b-mediated regulation occurs only during active myelination after P20 . To investigate the molecular basis of the observed myelin phenotype , we looked at the expression level of Dlg1 , a known interactor of Kif13b and a negative regulator of Schwann cell myelination in vivo [7–9] . Interestingly , we found that Dlg1 expression level was increased in Kif13bFl/Fl P0-Cre nerves at P20 ( Fig 3A ) . Note that the increase is particularly evident in the lower band ( Fig 3B ) , which corresponds to a hypo-phosphorylated isoform of Dlg1 [21 , 22] . In contrast , Dlg1 mRNA levels were downregulated in mutant nerves ( Fig 3C ) , which suggested that Dlg1 protein was more stable in the absence of Kif13b . To assess whether other negative regulators could contribute to the observed effect on myelination , we also looked at Ddit4/REDD1 expression levels . Ddit4/REDD1 is a known negative regulator of myelination , which downregulates the mechanistic target of rapamycin ( mTOR ) pathway by activating the tuberous sclerosis complex TSC1/2 [9] . We found that Ddit4 was similarly expressed between wild-type and mutant nerves at P10 and P20 ( Fig 3D ) . Dlg1 interacts with Kif13b in Schwann cells and is known to potentiate PTEN phosphatase activity on PIP3 , thus downregulating AKT activation [7–9] . Consistent with this , phosphorylation of AKT at S473 was decreased in Kif13bFl/Fl P0-Cre nerves as compared to controls at P20 , when AKT phosphorylation starts to decline during postnatal nerve development ( Fig 3E ) [9] . On the contrary , in Kif13bFl/Fl P0-Cre nerves , phosphorylation of AKT at T308 was not significantly different from controls ( Fig 3F ) . This finding may indicate activation of the feedback loop involving mTOR and molecules upstream of PI3K , as also already observed in other mutants [9 , 23–26] . Finally , we found normal expression levels of NRG1 type III ( and the phosphorylation of its receptor ErbB2 ) , Krox20 , and Oct6 , known regulators of myelin initiation , further supporting that reduced myelin thickness of Kif13bFl/Fl P0-Cre nerves is associated with enhanced negative regulation of postnatal myelination and not with a delay in myelin program initiation ( S2 Fig ) . Phosphorylation is known to modulate protein–protein interactions necessary for the cytoskeletal localization of Dlg1 [27 , 28] . In particular , serine phosphorylation correlates with Dlg1 inactivation , and hyperphosphorylated Dlg1 interacts with ubiquitin ligases , which mediate its ubiquitination and degradation [8 , 21 , 22 , 27–30] . Thus , we hypothesized that increased Dlg1 protein levels in Kif13bFl/Fl P0-Cre nerves could result from reduced serine phosphorylation and/or ubiquitination . By immunoprecipitating Dlg1 from sciatic nerve lysates at P20 , we observed a decrease of Dlg1-serine phosphorylation in Kif13bFl/Fl P0-Cre nerves compared to controls ( Fig 4A ) . As expected , in Dlg1Fl/Fl P0-Cre sciatic nerve lysates , the phosphorylated band was not detected ( Fig 4B ) . Then , to evaluate whether the decrease in serine-phosphorylation correlated with increased stability , we determined the pattern of Dlg1 ubiquitination . Consistent with our hypothesis , by immunoprecipitating Dlg1 from Kif13bFl/Fl P0-Cre nerves at P4 and P10 , we found that Dlg1 was less ubiquitinated in mutant nerve lysates when compared to controls ( Fig 4C ) . Our data suggest that the hypomyelination in Kif13bFl/Fl P0-Cre nerves results from increased Dlg1 stability/activity and enhanced negative regulation of AKT . Hence , we hypothesized that 50% reduction of Dlg1 gene expression in the Kif13bFl/Fl P0-Cre background might rebalance Dlg1 levels and rescue the phenotype . Thus , we generated Kif13bFl/Fl//Dlg1Fl/+; P0-Cre mice , and we compared these mutants with Kif13bFl/Fl//Dlg1+/+; P0-Cre mouse nerves . By performing western blot analysis , we observed that Dlg1 expression and AKT phosphorylation levels in Kif13bFl/Fl//Dlg1Fl/+; P0-Cre sciatic nerve lysates were rescued at a level similar to controls ( Fig 4E and 4F ) . Accordingly , myelin thickness in Kif13bFl/Fl//Dlg1Fl/+; P0-Cre nerves was also restored ( Fig 4D ) . Overall , our data suggest that Kif13b negatively regulates Dlg1 stability and activity in Schwann cells . Thus , in kif13bFl/Fl P0-Cre nerves , increased Dlg1 activity reduces AKT signaling and myelination . Since Kif13b regulates PNS myelination , we sought to assess whether Kif13b has a similar role in the CNS . First , we confirmed Kif13b mRNA expression in optic nerves and in myelinated tracts of the corpus callosum ( Fig 5A and 5B ) . Then , we generated a Kif13bFl/- CNP-Cre mouse with conditional inactivation of Kif13b in newly generated oligodendrocytes [31] . To achieve maximum efficiency of CNP-Cre mediated recombination , we generated a compound heterozygous mouse for a Kif13bFl allele and a Kif13b- ( null ) allele . We first assessed downregulation of Kif13b mRNA expression in Kif13b Fl/- CNP-Cre optic nerves by performing quantitative RT-PCR analysis ( Fig 5A ) . Western blot analysis confirmed reduction of Kif13b protein expression in lysates from corpus callosum of Kif13b Fl/- CNP-Cre mice ( Fig 5B ) . We then performed morphological analysis of optic nerves and spinal cords at P30 . Surprisingly , we observed increased myelin thickness with decreased g-ratios in both Kif13bFl/- CNP-Cre optic nerves and spinal cords as compared to either Kif13bFl/+ or Kif13b-/+ controls ( Fig 5C and 5D ) . However , at P90 , myelin thickness in either Kif13bFl/- CNP-Cre optic nerves or spinal cords was normal , suggesting a transient effect of Kif13b loss ( S3 Fig ) . At the molecular level , AKT phosphorylation at S473 was enhanced in both Kif13bFl/- CNP-Cre optic nerves and spinal cords at P30 ( Fig 5E ) , consistent with the observed hypermyelination and the role of AKT in promoting CNS myelination [32] . We then explored whether , as in the PNS , Kif13b regulates myelination by controlling Dlg1 expression levels . First , we assessed whether Kif13b interacts with Dlg1 in oligodendrocytes in vivo . By performing GST pull down assays from rat optic nerve lysates using GST-Kif13b/MBS as a bait , we identified Dlg1 , suggesting the existence of a Kif13b/Dlg1 complex ( Fig 5F ) . Interestingly , we noted that in spinal cord and optic nerve lysates Dlg1 isoforms were expressed in the range 140–150 KDa , as already observed in sciatic nerves ( Fig 3A ) , where Dlg1 is not expressed in the axon [7 , 8] . This finding suggests that the Kif13b/Dlg1 interaction likely occurs in oligodendrocytes and not in axons/neurons , where the main Dlg1/SAP97 isoform runs at a different molecular weight ( 97KDa ) . Next , we evaluated Dlg1 protein expression in Kif13bFl/- CNP-Cre mice and we found increased Dlg1 levels in both Kif13bFl/- CNP-Cre optic nerves and spinal cords at P30 ( Fig 5G ) . This result is consistent with the hypothesis that Kif13b negatively regulates Dlg1 expression also in the CNS . Overall , our findings indicate that Kif13b is a transient negative regulator of myelination in the CNS as its downregulation in oligodendrocytes increases myelin thickness and enhances AKT activation . Moreover , we suggest that also in the CNS Kif13b interacts with Dlg1 and negatively regulates its stability . In Kif13bFl/- CNP-Cre mice , increased myelin thickness is associated with enhanced Dlg1 expression . However , if Dlg1 acts as a negative regulator of myelination in oligodendrocytes as well , we would have expected to observe hypomyelination and not hypermyelination . Thus , we hypothesized that in oligodendrocytes Dlg1 might have the opposite role in the control of myelination , being a promoter rather than an inhibitor . To test this hypothesis , we generated Dlg1Fl/FlCNP-Cre conditional knockout mice in which Dlg1 was ablated in oligodendrocytes . We first demonstrated a reduction of Dlg1 protein expression in Dlg1Fl/Fl CNP-Cre optic nerves at P30 ( Fig 6A ) . Then , we performed morphological analyses of optic nerves and spinal cords starting at P30 . Consistent with our hypothesis , mutant optic nerves and spinal cords displayed reduced myelin thickness and increased g-ratios ( Fig 6C and 6D ) . Hypomyelination was also evidenced by decreased myelin basic protein ( MBP ) expression levels in spinal cord lysates from Dlg1Fl/Fl CNP-Cre mice ( Fig 6B ) . As in the case of Kif13bFl/- CNP-Cre mutants , myelin thickness of Dlg1Fl/Fl CNP-Cre optic nerves and spinal cords was normal at P90 , suggesting a transient role of Dlg1 in the control of myelination ( S3 Fig ) . To investigate the mechanism by which Dlg1 promotes myelination in oligodendrocytes , we examined the phosphorylation state of AKT in lysates from optic nerves and corpus callosum of Dlg1Fl/Fl CNP-Cre mutants . We found that AKT phosphorylation at both S473 and T308 was reduced in both Dlg1Fl/Fl CNP-Cre optic nerves and corpus callosum as compared to controls , consistent with the decreased myelination ( Fig 6E and 6F ) . Since ( 1 ) AKT phosphorylation depends on PIP3 levels and on the activity of the PI3K class I and ( 2 ) Dlg1 has been described to interact with the regulatory subunit of PI3K class I , p85 , in epithelial cells [29] , we hypothesized that also in oligodendrocytes Dlg1 may interact with p85 , influencing PI3K activity upstream of AKT . To address this point , we first explored p85 expression levels in optic nerves and spinal cords at P30 and found that p85 protein levels were reduced in Dlg1Fl/Fl CNP-Cre mice ( Fig 6H–6I' ) . Next , GST pull down experiments from P11 rat optic nerve lysates demonstrated that Dlg1 and p85 are interactors of GST-Kif13b/MBS ( Fig 6G ) , thus providing evidence for the existence of a complex involving Kif13b , Dlg1 , and p85 . Interestingly , by performing co-immunoprecipitation and pull down experiments , we did not observe interaction between p85 and the Kif13b/Dlg1 complex in the PNS in sciatic nerves . Consistent with this , p85 was similarly expressed in Kif13bFl/Fl P0-Cre and Dlg1Fl/Fl P0-Cre mutant sciatic nerves as compared to controls ( S4 Fig ) . These findings suggest that in the PNS , in contrast to the CNS , the Kif13b/Dlg1 complex does not involve p85 . In conclusion , similarly to Schwann cells , downregulation of Kif13b expression in oligodendrocytes is associated with increased Dlg1 levels . However , in the CNS , Dlg1 promotes myelination . Thus , downregulation of Kif13b expression in oligodendrocytes causes hypermyelination . Finally , we asked how downregulation of Kif13b expression results in increased Dlg1 stability in both PNS and CNS . In previous yeast two-hybrid screening analyses , we had found that the PDZ2+3 domain of Dlg1 directly interacts with the p38γ MAPK isoform [7 , 33] , as also previously reported for HEK293 cells [28] . Since p38γ can phosphorylate and negatively regulate the interaction of Dlg1 with cytoskeletal protein partners , we further investigated the interaction of Kif13b , p38γ , and Dlg1 in the nerve in vivo . We first confirmed Dlg1 and p38γ interaction by performing co-immunoprecipitation experiments from sciatic nerve lysates ( Fig 7A ) . Next , we observed that Kif13b/MBS-GST was able to pull down both Dlg1 and p38γ from nerve lysates , suggesting that Kif13b , p38γ , and Dlg1 may be part of the same complex ( Fig 7B ) . To provide further evidence for this hypothesis , we investigated p38γ expression levels in mutants with conditional ablation of either Kif13b or Dlg1 in Schwann cells . Interestingly , p38γ expression levels were decreased in Kif13bFl/Fl P0-Cre sciatic nerves at both P20 and 9 mo ( Fig 7C and 7D ) but not in Dlg1Fl/Fl P0-Cre nerves ( Fig 7E and 7F ) , suggesting that p38γ acts downstream of Kif13b and upstream of Dlg1 . To confirm these results , we analyzed the sciatic nerves of p38γ knock-out mutants . As expected , nerves from p38γ-null mice were hypomyelinated ( Fig 7G–7I ) , supporting the hypothesis that p38γ is a novel promoter of Schwann cell myelination . Finally , since our data suggest that Kif13b may similarly regulate Dlg1 also in the CNS , we assessed whether a Kif13b , p38γ , and Dlg1 complex could be detected in oligodendrocytes . As expected , Dlg1 and p38γ co-immunoprecipitate from optic nerve lysates ( Fig 8A ) and Kif13b/MBS-GST is able to pull down both Dlg1 and p38γ ( Fig 8B ) . Even if not as striking as in Schwann cells , p38γ expression levels were decreased in Kif13bFl/- CNP-Cre optic nerve lysates but not in Dlg1Fl/Fl CNP-Cre , suggesting that p38γ acts downstream of Kif13b and upstream of Dlg1 ( Fig 8C–8E ) . As p38α is the MAPK isoform known to regulate myelination in both PNS and CNS [34–41] , we assessed whether Kif13b/Dlg1 may also form a complex with p38α . Interestingly , by performing pull down experiments , we found that Dlg1 does not interact with p38α in either optic or sciatic nerve lysates . Consistent with this , expression levels of p38α in either sciatic nerves or spinal cords of Kif13b conditional knock-out mutants were similar to controls ( S5 Fig ) . Overall , these findings suggest a similar mechanism of Kif13b and p38γ-mediated regulation of Dlg1 in both PNS and CNS , with opposite outcomes on the control of myelination , as Dlg1 is a brake on myelination in the PNS and a positive regulator in the CNS .
Here we report that downregulation of Kif13b expression in Schwann cells is associated with reduced myelin thickness , decreased AKT activation , and increased levels of Dlg1 , a known brake on PNS myelination acting on the PIP3-AKT-mTOR pathway [8 , 9] . As Kif13b and Dlg1 interact in Schwann cells [7] , we hypothesized that Kif13b may control myelination through the Dlg1 scaffold itself , by regulating its stability and function . Indeed , in support of our hypothesis , in Kif13bFl/Fl//Dlg1Fl/+; P0-Cre double mutant sciatic nerves , Dlg1 expression levels and myelin thickness are similar to wild type . Interestingly , we report here that in Kif13bFl/Fl P0-Cre nerves , in which Dlg1 expression levels are increased , myelination is not delayed at very early stages of postnatal nerve development , and reduced myelin thickness is evident when AKT activation starts to physiologically decline , after P20 [8 , 9] . This observation is consistent with the phenotype of mutant mice lacking Dlg1 , specifically in Schwann cells [9] . We previously reported a transient increase in myelin thickness and occasional myelin outfoldings in Dlg1Fl/Fl P0-Cre nerves starting from P10 [9] . However , even if enhanced , myelination was not accelerated in Dlg1Fl/Fl P0-Cre nerves , and , at very early stages of postnatal nerve development , the number of myelinated fibers and myelin thickness were similar to control nerves . Thus , Dlg1 may act as a brake on myelination to downregulate AKT activation at the peak of myelination , when AKT phosphorylation starts to decline . In support to this hypothesis , myelin outfoldings , a focal form of hypermyelination that is thought to be linked to AKT overactivation and loss of Dlg1-mediated negative control on myelination , are observed in the nerve after 3 w of postnatal development [44] . Dlg1 stability is controlled by phosphorylation and ubiquitination [8 , 21 , 22 , 27 , 28 , 30 , 45 , 46] . In Drosophila , the PAR1 kinase directly phosphorylates Dlg1 at conserved sites and negatively regulates its mobility and targeting at postsynaptic membranes of neuromuscular junctions [27] . Osmotic stress-induced serine phosphorylation of Dlg1 by p38γ MAP kinase can induce Dlg1 dissociation from the glucokinase-associated dual specificity phosphatase ( GKAP ) and the cytoskeleton , negatively regulating Dlg1 [28] . Finally , phosphorylated DLG1 interacts with the β-TrCP ubiquitin ligase receptor , which mediates ubiquitination of the protein [30] . Thus , we investigated whether enhanced Dlg1 protein expression levels in Kif13bFl/Fl P0-Cre nerves correlated with a decrease in serine phosphorylation and/or ubiquitination . Consistent with our hypothesis , we found that in Kif13bFl/Fl P0-Cre nerves Dlg1 is less phosphorylated and less ubiquitinated , suggesting that Kif13b promotes radial myelin growth by directly or indirectly influencing Dlg1 stability and expression . We also suggest that p38γ MAPK could be the kinase that , downstream of Kif13b , phosphorylates Dlg1 to regulate its activity . Indeed , p38γ MAPK is known to interact with and to phosphorylate serine residues of Dlg1 in other cells [28] . We identified p38γ in a yeast two-hybrid screening analysis using a nerve cDNA library and Dlg1 as a bait [7 , 33] . Moreover , we show that p38γ , Dlg1 , and Kif13b form a complex in the nerve . More importantly , sciatic nerves of p38γ-null mice are hypomyelinated , thus confirming the hypothesis that p38γ , by phosphorylating and negatively regulating Dlg1 , acts as a promoter of myelination downstream of Kif13b . Unfortunately , antibodies that can specifically recognise phosphorylated p38γ are not available to assess whether activated p38γ could interact with Kif13b and Dlg1 . Interestingly , the role of p38γ MAPK in the regulation of PNS myelination has not yet been assessed . Previous studies suggested that p38 MAPK mediates laminin signaling in vitro to promote Schwann cell elongation and alignment at the very first stages of differentiation [34] . Hossain et al . , suggested that p38 directs Schwann cell differentiation by regulating Krox-20 expression , thus further supporting the role of p38 MAPK as a positive regulator of PNS myelination [35] . However , on the basis of the MAPK inhibitors used , the observed effect was likely to be mediated by the p38α or p38β [35] . A more recent study reported that in vitro p38 MAPK promotes the de-differentiated state of Schwann cells during Wallerian degeneration , by inducing c-Jun expression and by inhibiting myelin gene expression , and also suggested that p38 MAPK is a negative regulator of Schwann cell differentiation and myelination during development [36] . On the basis of the antibodies used recognizing the phosphorylated state of MAPK as well as the MAPK inhibitor used ( SB203580 ) , other isoforms rather than p38γ are more likely to mediate this function [36] . How can both Kif13b and p38γ control Dlg1 phosphorylation , ubiquitination , and stability ? Kif13b could transport and localize the kinase at membranes where Dlg1 is enriched to downregulate , in complex with PTEN , PIP3 levels , and AKT activation [47] . Indeed , in Kif13b-null but not in Dlg1-null nerves p38γ expression levels are reduced , thus suggesting that p38γ is downstream of Kif13b and upstream of Dlg1 . Alternatively , the binding of Kif13b with Dlg1 , which is mediated by the membrane-associated guanylate kinase homologue binding stalk ( MBS ) and guanylate kinase homologue ( GUK ) domains , respectively , may relieve intramolecular inhibition in either Kif13b or Dlg1 , as already reported [48] . For example , following Kif13b binding , a conformational change in Dlg1 ( open state ) can be induced so that target residues for serine phosphorylation can be exposed and accessible to p38γ kinase-mediated phosphorylation . Unfortunately , p38γ-specific inhibitors are not available to further investigate these mechanisms . Our data convey a novel function for Kif13b/p38γ as negative regulators of Dlg1 in the PI3K/AKT signalling pathway . Interestingly , Kif13b has already been proposed as a negative regulator in other studies . For example , in PC12 cells , KIF13B negatively regulates centaurin-α1/PIP3BP ( PIP3 binding protein ) , a GAP for Arf6 , thus promoting Arf6 GTPase plasma membrane activation [16] . Further , in T cells , KIF13B negatively regulates TCR signaling to NF-kB , by redistributing the CARD11 scaffold from the center of the synapse to a more distal region [13] . We also show that Kif13b is a negative regulator of CNS myelination . Indeed , we observed that downregulation of Kif13b expression in oligodendrocytes results in increased myelin thickness and AKT activation , consistently with the role of AKT in promoting CNS myelination [32] . Similar to PNS , we found that Kif13b interacts with Dlg1 and that loss of Kif13b is associated with increased Dlg1 levels , thus suggesting a negative regulation mediated by Kif13b on Dlg1 . Given these similarities , we investigated whether the increased Dlg1 level and stability in oligodendrocytes could also result from a decrease in p38γ-mediated phosphorylation . Indeed , we found that Kif13b , Dlg1 , and p38γ MAPK interact in optic nerves and that p38γ expression is decreased in Kif13b but not in Dlg1 mutants , as already observed in the PNS . These findings suggest that p38γ may act downstream of Kif13b and upstream of Dlg1 to negatively regulate Dlg1 activity . The role of the p38γ isoform in the regulation of CNS myelination has not been yet assessed . As for PNS , only p38α has been investigated in the CNS . Inhibition of p38α activity or expression in vitro in a co-culture system has been reported to prevent oligodendrocyte progenitor differentiation and myelination [37–39] . Another study suggested that p38α MAPK supports myelin gene expression in the brain through several mechanisms acting on both positive and negative regulators of differentiation [40] . More recently , myelination was found to be impaired in mice with conditional inactivation of p38α MAPK in oligodendrocyte progenitor cells [41] . Interestingly , the same authors observed an opposite effect of p38α MAPK in remyelination , as mutant mice exhibited a more efficient remyelination as compared to controls following demyelination [41] . These studies further support the notion that the regulation of myelination is a very complicated process , in which different signals arising from the extracellular matrix , axons , and astrocytes in the CNS must be correctly integrated in time and space within the same cell to achieve homeostasis [49–56] . If Dlg1 is a brake on myelination in the CNS as in the PNS , how can loss of Kif13b and elevation of Dlg1 result in increased CNS myelin thickness ? Surprisingly , our data indicate that in oligodendrocytes Dlg1 is a positive and not a negative regulator of myelination , as its loss is associated with reduced myelin thickness and AKT activation . Interestingly , in addition to Dlg1 , other molecules have been found to control myelination with opposite roles in PNS and CNS [57–60] . For example , myosin light chain II phosphorylation promotes myelination in the PNS and inhibits myelination in the CNS [57] . To determine the mechanism by which Dlg1 could promote CNS myelination acting on the PI3K-AKT pathway , we sought to investigate the regulatory subunit of PI3K class I , p85 , a known interactor of Dlg1 in epithelial cells [29] . Consistent with this , we found that Dlg1 interacts with p85 in the optic nerve , likely to modulate PI3K class I activity , PIP3 levels , and ultimately AKT activation . Interestingly , phosphorylation of DLG1 on serine and threonine is known to prevent DLG1 interaction with SH2 domains of p85/PI3K [29] . Thus , we could speculate that Dlg1 , when hypophosphorylated , may display a higher affinity for the SH2 domains of p85 , whose activation is necessary for PI3K activity regulation [61] . Whether in oligodendrocytes Dlg1 also promotes myelination by other mechanisms , which can converge on AKT activation , remains to be determined .
The following primary antibodies were used: mouse anti-KIF13B ( provided by Dr . A . Chishti ) ; mouse anti-Dlg1 ( Stressgen ) ; mouse anti-phosphoserine ( Alexis Biochemicals ) ; rabbit anti-ubiquitin ( Santa Cruz Biotechnology ) ; rabbit anti-DRP2 ( provided by Dr . D . Sherman ) ; rabbit anti-phospho-Akt ( Ser473 ) ( Cell Signaling ) ; rabbit anti-phospho-Akt ( Thr308 ) ( Cell Signaling ) ; rabbit anti-Akt ( pan ) ( Cell Signaling ) ; rabbit anti-phospho-p44/42 MAPK ( Erk1/2 ) ( Thr202/Tyr204 ) ( Cell Signaling ) ; rabbit anti-p44/42 MAP Kinase ( Cell Signaling ) ; rabbit anti-Neuregulin-1α/β1/2 ( C20 ) ( Santa Cruz Biotechnology ) ; rabbit anti-p-Neu ( Tyr 1248 ) -R ( i . e . , p-ErbB-2 ) ( Santa Cruz Biotechnology ) ; rabbit anti-Neu ( C-18 ) ( i . e . , ErbB-2 ) ( Santa Cruz Biotechnology ) ; rabbit anti-PI3 Kinase p85 ( Cell Signaling ) ; rat anti-MBP ( Millipore ) ; rabbit anti-p38α ( Santa Cruz ) ; rabbit anti-p38γ ( R&D Systems ) ; rabbit anti-calnexin ( Sigma-Aldrich ) ; mouse anti-β-tubulin ( Sigma-Aldrich ) ; rabbit anti-actin ( Sigma-Aldrich ) . For immunofluorescence , secondary antibodies included fluorescein ( FITC ) -conjugated and rhodamine ( TRITC ) -conjugated donkey anti-mouse or rabbit IgG ( Jackson ImmunoResearch ) . For western blotting , secondary antibodies included horseradish peroxidase ( HRP ) -conjugated goat anti-rabbit and rabbit anti-mouse immunoglobulins ( Dako ) , and IRDye 800- and 680-conjugated goat anti-mouse , goat anti-rabbit , and goat anti-rat IgG ( Li-Cor Biosciences ) . As negative control in immunoprecipitation experiments , ChromPure mouse IgG whole molecules were used ( Jackson ImmunoResearch ) . All experiments involving animals were performed in accordance with Italian national regulations and covered by experimental protocols reviewed by local Institutional Animal Care and Use Committees . The pFlrt-1 vector , including lox-P sites , FRT-flanked neomycin resistance gene ( neo ) , and PGK-TK , was used to target the Kif13b gene . The selected Kif13b mouse genomic regions to be inserted in the targeting vector were amplified from a BAC clone spanning the Kif13b gene and obtained from The Center for Applied Genomics ( The Hospital for Sick Children , Ontario , Canada ) . To generate the targeting vector for homologous recombination , a 503 bp BamHI fragment including exon 6 and flanking intronic regions was first inserted between lox-P sites in pFlrt-1 . In a second step , a 4 , 606 bp fragment containing exon 7 was inserted into the BstBI site downstream of the PGK-neo cassette to constitute the long arm for homologous recombination . Finally , a fragment of 2 , 000 bp containing exon 5 was cloned into the SalI site upstream to the first lox-P and represented the short arm for homologous recombination . After electroporation of TBV2 embyonic stem cells ( 129S2/SvPas ) , recombinant clones were screened by Southern blot analysis . Digestion with KpnI and hybridization with two probes designed on exon 6 ( inside the recombination ) and upstream of exon 5 ( outside the 5′ end of the recombination ) revealed two bands of 7 , 671 bp ( wild type ) and of 9 , 671 bp ( containing the neo cassette ) . Similarly , SmaI digestion of genomic DNA and hybridization using a probe designed at the 3′ end of the targeted region , outside the recombination boundaries , detected two bands of 7 , 694 bp ( the targeted allele , since one SmaI restriction site is present within the neo cassette ) and of 9 , 151 bp ( the wild-type allele ) . Two different correctly targeted clones were injected into C57BL6 blastocysts ( Core Facility for Conditional Mutagenesis San Raffaele/Telethon Transgenic Service ) to obtain transmission of the Floxed allele through the germline . The neo cassette was removed in vivo by crossing heterozygous Kif13bFl ( neo ) /+ with Flpe transgenic mice . Heterozygous Kif13bFl/+ animals were crossed with P0-Cre [17 , 18] transgenic mice to excise exon 6 specifically in Schwann cells . To generate Kif13bFl/Fl P0-Cre conditional knockout mice , Kif13bFl/+ P0-Cre animals were crossed with homozygous Kif13bFl/Fl . Kif13bFl/Fl mouse nerves had normal myelin thickness and mean g-ratio values similar to wild-type mice , thus suggesting that Kif13bFl/+ does not represent a hypomorphic allele . To obtain Kif13bFl/- CNP-Cre [31] mice with conditional inactivation of Kif13b in oligodendrocytes , Kif13bFl/+ mice were first crossed with CMV-Cre transgenic mice . Then , after germline segregation of the CMV-Cre transgene , Kif13b -/+ ( without CMV-Cre ) were crossed with Kif13bFl/+ CNP-Cre mice to obtain Kif13bFl/- CNP-Cre conditional null . In this way , we increased CNP-Cre mediated recombination efficiency on the Floxed allele in the Kif13bFl/- CNP-Cre genotype . The Dlg1Fl ( C57/BL6 strain ) allele has been already reported ( Zhou et al . , 2008 ) . To generate Dlg1 conditional knockout mice in oligodendrocytes , homozygous Dlg1Fl/Fl mice were crossed with heterozygous Dlg1Fl/+ mice carrying the CNP-Cre transgene . To obtain 50% reduction of Dlg1 specifically in Schwann cells in a Kif13bFl/Fl P0-Cre background , Kif13bFl/Fl P0-Cre mice were first crossed with Dlg1Fl/Fl P0-Cre mice . Then , Kif13bFl/+//Dlg1Fl/+; P0-Cre double heterozygous mice were crossed to obtain Kif13bFl/Fl//Dlg1Fl/+; P0-Cre mice . These latter were compared with Kif13bFl/Fl//Dlg1+/+; P0-Cre mice and controls ( only floxed alleles without Cre ) within the same litters . The generation of p38γ-null mice has been already reported [28] . For all the experiments involving animals , n ≥ 5 animals per genotype of either sex were analysed . Genotype analysis on Kif13b mutant mice was carried out on tail genomic DNA using primer pairs A plus B ( 415 bp floxed band and 342 bp wild type band ) or A plus C ( 966 bp floxed band , 834 bp wild type band , and 378 bp recombined band ) . Genotype analysis of the Dlg1 floxed allele and of the p38γ-null locus has already been reported [9 , 28] . RT-PCR was performed as described previously [7 , 9] . Designed probes were used to amplify mouse Kif13b and the endogenous reference transcript calnexin . The comparative Ct method was used . As calibrator , a control sample ΔCt was chosen for each selected transcript . The ΔΔCt ( ΔCt of each normalized selected transcript minus ΔCt of the calibrator ) was calculated . Expression levels of Kif13b mRNA are indicated as 2-ΔΔCt values . For statistical analysis , SD was calculated for triplicate samples of each reaction and SEM is indicated on the average of the determinations from different animals . Three to five animals per genotype for each time point were analysed . Semithin analysis of quadriceps and sciatic nerves and ultrastructural analysis of optic nerves and spinal cords were performed as described previously [62] . To perform morphometric analysis , digitalized images of fiber cross sections were obtained from corresponding levels of the quadriceps or sciatic nerves with a 100x objective and Leica DFC300F digital camera ( Milan , Italy ) . Five images per animal were analysed with the Leica QWin software ( Leica Microsystem ) and the g-ratio calculated as the ratio between the mean diameter of an axon ( without myelin ) and the mean diameter of the same axon including the myelin sheath . For morphometric analysis on ultrastructural sections , 20 images per animal were taken at 4000x ( LEO 912AB Transmission Electron Microscope , Milan , Italy ) and the g-ratio values determined by measuring axon and fiber diameters . Internodal lengths were measured as described using Openlab ( PerkinElmer ) [19] , and 100 internodes of two quadriceps nerves were evaluated for each animal ( n = 3 ) . Adult mice were anesthetized with avertin ( trichloroethanol , 0 . 02 ml/g of body weight ) , and crush injury was performed as previously described [63] . After skin incision , the sciatic nerve was exposed and crushed distal to the sciatic notch for 20 s with fine forceps previously cooled in dry ice . To identify the site of injury , forceps were previously dropped into vital carbon . The nerve was replaced under the muscle and the incision sutured . Protein lysates from mouse sciatic nerves , corpus callosum , optic nerves , and spinal cords for western blot analysis were prepared using a lysis buffer containing 2% SDS , 50 mM Tris buffer pH 8 . 0 , 150 mM NaCl , 10 mM NaF , 1 mM NaVO3 , and complete protease and phosphatase inhibitors ( Roche ) . For the detection of phosphorylated antigens , samples were lysed with a buffer containing 1%TX-100 . Protein quantification was performed using BCA assay ( Pierce , Thermo Scientific ) . Mouse and rat sciatic nerves were lysed in a buffer containing 1% NP-40 , 150 mM NaCl , 50 mM Tris buffer pH 8 . 0 , 10 mM NaF , 1 mM NaVO3 , and complete protease and phosphatase inhibitors ( Roche ) . Following centrifugation at 13 , 000 rpm for 15 min at 4°C , equal amounts of protein lysates were incubated with 6–8 ug of mouse anti-Dlg1 antibody ( Stressgen ) or mouse IgG for control ( Jackson ImmunoResearch ) . After 3 h of incubation with the antibody at 4°C , 35 μl of protein G agarose ( settled ) ( Sigma-Aldrich ) was added to immunocomplexes within the lysates and incubated for 1 h and 30 min at 4°C . The agarose beads were washed two times with cold PBS-Tween 0 . 1% and once with cold PBS . The immunoprecipitated product was denatured in Laemmly buffer ( Biorad ) with β-mercaptoethanol and resolved by SDS-PAGE . Kif13b/MBS cDNA was cloned into pGEX-4T2 expression vector and expressed together with GST alone in Escherichia coli BL21 ( DE3 ) cells [7] . Recombinant proteins were purified directly from bacterial extract on glutathione-Sepharose 4 Fast Flow beads . Rat sciatic and optic nerves were lysed in a buffer containing 1% NP-40 , 50 mM Tris buffer pH 7 . 4 , 10% glycerol , 100 mM NaCl , 10 mM NaF , and 1 mM NaVO3 . Equal amounts of protein lysates were incubated for 4 h at 4°C with immobilized GST-Kif13b/MBS proteins and GST as control . After three washes with a buffer containing 0 . 5% NP-40 , 50 mM Tris buffer pH 7 . 4 , 10% glycerol , 100 mM NaCl , 10 mM NaF , and 1 mM NaVO3 , the bead pellets were dissolved in Laemmly buffer with β-mercaptoethanol , resolved by SDS-PAGE , and analyzed by immunoblotting . To show the relative amount used of GST-Kif13B/MBS and GST , beads were dissolved again in Laemmly buffer with β-mercaptoethanol , resolved by SDS-PAGE , and the gels stained with Coomassie . SDS-PAGE gels were transferred to PVDF membranes ( Millipore ) or to nitrocellulose ( Millipore ) at 4°C in 20% methanol blotting buffer . Filters were blocked in 5% dry milk in PBS-0 . 1% Tween 20 overnight at 4°C and immunoblotted with primary antibodies diluted in 3% dry milk in PBS-0 . 1% Tween . For phosphorylated antigens , an additional blocking was performed for 30 min at RT in 3% bovine serum albumin ( BSA ) ( Sigma-Aldrich ) , 0 . 5% gelatin , 0 . 1% Tween , 1 mM EDTA pH 8 . 0 , 0 . 15 M NaCl , 10 mM Tris buffer pH 7 . 5 , followed by incubation with primary antibodies diluted in the same blocking solution . Secondary antibodies , either horseradish peroxidase-conjugated ( Dako ) or IRDye 800- and 680-conjugated ( Li-Cor Biosciences ) , were used and immunoblots revealed by using either ECL/ECL-prime developing systems and films for chemiluminescent detection ( Amersham ) or by Odyssey CLx Infrared Imaging System ( Li-Cor Biosciences ) . Statistical analysis was performed using the Student t test , two tails , unequal variants , and α = 0 . 005 were considered . All results are shown as mean ± SEM . Figures were prepared using Adobe Photoshop version 11 . 0 ( Adobe Systems ) . | Myelin is a multilayered extension of the Schwann and oligodendrocyte cell membranes , which wraps around neuronal axons to facilitate propagation of electric signals and to support axonal metabolism . However , the signals regulating myelin formation and how they are integrated and controlled to achieve homeostasis are still poorly understood . In Schwann cells , the Discs large 1 ( Dlg1 ) protein is a known brake of myelination , which negatively regulates the amount of myelin produced so that myelin thickness is proportional to axonal diameter . In this paper , we report that in Schwann cells Dlg1 itself is tightly regulated to ensure proper myelination . We propose that Dlg1 function is further controlled by the Kif13b kinesin motor protein , which acts as a "brake of the brake" by downregulating Dlg1 activity . Surprisingly , we found that in oligodendrocytes Dlg1 is a positive and not a negative regulator of myelination . Thus , Kif13b-mediated negative regulation of Dlg1 ensures appropriate myelin production and thickness in the central nervous system . Our data further extend recently emerged unconventional roles for kinesins , which are usually implicated in cargo transport rather than in the modulation of signaling pathways . The elucidation of mechanisms regulating myelination may help to design specific approaches to favor re-myelination in demyelinating disorders in which this process is severely impaired . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] | [
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"nervous",
"system",
"neuroscience",
"macroglial",
"cells",
"optic",
"nerve",
"schwann",
"cells",
"nerve",
"fibers",
"spinal",
"cord",
"animal",
"cells",
"proteins",
"glial",
"cells",
"biochemistry",
"cellula... | 2016 | Kif13b Regulates PNS and CNS Myelination through the Dlg1 Scaffold |
Alveolar echinococcosis , caused by Echinococcus multilocularis larvae , is a chronic disease associated with considerable modulation of the host immune response . Dendritic cells ( DC ) are key effectors in shaping the immune response and among the first cells encountered by the parasite during an infection . Although it is assumed that E . multilocularis , by excretory/secretory ( E/S ) -products , specifically affects DC to deviate immune responses , little information is available on the molecular nature of respective E/S-products and their mode of action . We established cultivation systems for exposing DC to live material from early ( oncosphere ) , chronic ( metacestode ) and late ( protoscolex ) infectious stages . When co-incubated with Echinococcus primary cells , representing the invading oncosphere , or metacestode vesicles , a significant proportion of DC underwent apoptosis and the surviving DC failed to mature . In contrast , DC exposed to protoscoleces upregulated maturation markers and did not undergo apoptosis . After pre-incubation with primary cells and metacestode vesicles , DC showed a strongly impaired ability to be activated by the TLR ligand LPS , which was not observed in DC pre-treated with protoscolex E/S-products . While none of the larvae induced the secretion of pro-inflammatory IL-12p70 , the production of immunosuppressive IL-10 was elevated in response to primary cell E/S-products . Finally , upon incubation with DC and naïve T-cells , E/S-products from metacestode vesicles led to a significant expansion of Foxp3+ T cells in vitro . This is the first report on the induction of apoptosis in DC by cestode E/S-products . Our data indicate that the early infective stage of E . multilocularis is a strong inducer of tolerance in DC , which is most probably important for generating an immunosuppressive environment at an infection phase in which the parasite is highly vulnerable to host attacks . The induction of CD4+CD25+Foxp3+ T cells through metacestode E/S-products suggests that these cells fulfill an important role for parasite persistence during chronic echinococcosis .
The metacestode larval stage of the fox-tapeworm E . multilocularis is the causative agent of alveolar echinococcosis , one of the most dangerous zoonoses world-wide [1] . Apart from the strobilar adult stage that resides within the intestine of the definitive host ( e . g . foxes , dogs ) , the life cycle of this cestode comprises three larval stages that are involved in the infection of the intermediate host ( small rodents and , occasionally , humans ) . An infection of the intermediate host is initiated by the oral uptake of ‘infectious eggs’ that contain the first larval stage , the oncosphere . Upon activation within stomach and intestine , the oncosphere hatches , penetrates the intestinal wall , and gains access to the host's viscera . Almost exclusively within the intermediate host's liver , the oncosphere then undergoes a metamorphosis towards the metacestodes which is driven by totipotent parasite stem cells ( germinal cells; neoblasts ) that were carried into the host through the oncosphere . As a result of the oncosphere - metacestode metamorphosis , fully mature , cyst-like metacestode vesicles are formed that grow infiltratively , like a malignant tumor , into the surrounding host tissue and that consist of an inner , cellular ‘germinal layer’ ( GL ) and an outer , glycan-rich and acellular ‘laminated layer’ ( LL ) [2] . At least in experimentally infected mice , the formation of the LL cannot be observed earlier than 2–3 weeks upon initial infection [3] , [4] , [5] , [6] . Evidence has been obtained that the LL is one of the parasite's key structures for protection against the host immune system in the later phase of the infection [7] . Approximately 2 months after the infection of mice , ‘brood-capsules’ are formed from stem cells of the GL that later give rise to the third larval stage , the protoscolex , which is passed on to the definitive host [2] . The E . multilocularis infection process can thus be separated into 3 phases . The first phase starts with the oncosphere and culminates , after 2–3 weeks , in the formation of mature metacestode vesicles which , in the second phase , grow infiltratively into the host tissue . During the third stage , protoscoleces are formed in natural intermediate hosts , but only rarely in human infections [2] . Cellular effector mechanisms are considered to be the key defense against metacestode growth and dissemination in mice and humans since mouse strains that cannot develop cellular immune responses are highly susceptible to AE , whereas strains defective in humoral immunity can control parasite growth to a certain level [8] . Furthermore , in humans co-infected with E . multilocularis and the human immunodeficiency virus ( HIV ) , very fast and unlimited parasite proliferation occurs [9] , [10] , whereas promotion of cellular immunocompetence has a beneficial effect on the outcome of the disease [8] . A significant number of studies on both humans and mice indicated that T helper 1 ( Th1 ) -dominated immune responses , characterized by the release of interferon-γ ( IFN-γ ) , after priming by DC that secrete interleukin ( IL ) -12 , are effective in eliminating the parasite at an early stage , whereas a Th2 cytokine profile ( IL-4 , IL-5 ) and the release of immunosuppressive IL-10 and TGF-β is generally associated with susceptibility to the parasite and progressive disease [8] . Although it became clear from these studies that the parasite , most probably by E/S-products , actively influences the host immune response ( e . g . gradually driving it into the Th2 branch ) , little is currently known on the molecular and cellular basis of E . multilocularis induced immune suppression , particularly for the early stages of the infection . DC are professional antigen presenting cells that represent the link between the innate and the adaptive immune system and are crucially involved in the induction of Th1- , Th2- or Th17-dominated immune responses [11] , [12] . Upon pathogen recognition , DC take up antigens and undergo maturation , as can be assessed by the up-regulation of surface markers such as the major histocompatibility complex II ( MHC II ) and co-stimulatory molecules such as CD86 and CD80 [11]–[13] . After migration to the T cell area of lymph nodes , DC interact with naïve T cells to promote adaptive immune responses towards the Th1- , Th2- , Th17-branches , depending on the pathogen pattern they have adopted [13] . However , DC are also targets of parasites to establish immune evasion , e . g . by induction of regulatory T cells ( T-reg ) , which counteract T helper cell activities [11] , [12] , [14] . A potentially important role of DC in immunosuppressive mechanisms during AE has indeed been established in a recent in vivo study on secondary ( intraperitoneal ) AE in mice [15] . In this work , Mejri et al . demonstrated that peritoneal DC from chronically infected mice , representing the late stage of AE , express higher levels of TGF-β mRNA , lower levels of IL-10 and IL-12 mRNA , and display down-regulation of maturation-associated surface markers , when compared to DC from non-infected mice [15] . Furthermore , DC from intraperitoneally infected mice specifically modulated CD4+ and CD8+ T cell responses suggesting a role for immunosuppressive T-reg during chronic AE [15] . DC are also among the first cells encountered by parasites during an infection [11] , [13] and may have a critical role in the Th1 to Th2 shift reported for the intermediate host during the chronic phase of AE [8] , [10] . In line with this hypothesis are recent data demonstrating that immature human DC fail to mature in the presence of crude , non-fractionated E . multilocularis antigen [16] . Moreover , during infection of the intermediate host , migration of parasitic larvae from the intestinal entry site to the liver and late metastasis to other organs ( lung , brain ) [17] strongly suggest that these larvae encounter DC in vivo . However and in spite of the general importance of DC in cellular host-helminth interaction mechanisms [11] , [12] , [18] , only few investigations have so far been carried out towards an identification and characterization of immunomodulatory molecules that are released by Echinococcus larvae and their influence on DC function . Apart from the above mentioned study concerning the influence of crude E . multilocularis antigen on human DC [16] , there are merely reports on the activity of crude hydatid ( vesicle ) fluid or selected hydatid fluid protein compounds of the related tapeworm E . granulosus on DC maturation [19] , [20] . Due to the limited availability of respective parasite material ( oncospheres ) , no in vitro studies have so far been carried out concerning the interaction of host immune cells with early infective parasite stages . Our current picture concerning the effects of Echincoccus E/S-products on host cells thus mostly derives from studies in which easier accessible protoscoleces had been employed [21]–[27] , with the considerable drawback that this stage is formed very late during an infection of the intermediate host ( if at all ) , and that in intact metacestode vesicles , protoscoleces do not have direct contact to host tissue and cells . Notably , we have recently introduced a primary germinal cell cultivation system by which the early developmental phase within the intermediate host can be re-constituted in vitro [28] . In this system , isolated E . multilocularis primary cells proliferate and form cellular aggregates that give rise to mature metacestode vesicles ( including LL ) in a manner that closely mimics the natural oncosphere - metacestode metamorphosis process [28] . Even concerning gene and protein expression patterns , this system closely reflects early parasite development within the intermediate host , and parasite antigens originally described to be expressed specifically in the oncosphere are readily detectable in the regenerating parasite cell aggregates [29] , [30] . Although it became clear from previous studies that E . multilocularis through its larval E/S-products tightly down regulates accessory cell functions of macrophages [31] little is currently known about the effect on DC . In the present study , we used our primary germinal cell cultivation system [28] to investigate the influence of E/S-products from primary cells ( characteristic of the early phase of the infection ) on DC and compared it with the effects of E/S-products of mature metacestode vesicles , characteristic for the chronic phase , and protoscoleces which , in intact parasite material , do not have direct contact with host immune cells .
All experiments were carried out in accordance with European and German regulations on the protection of animals ( Tierschutzgesetz ) . Ethical approval of the study was obtained from the local ethics committee of the government of Lower Franconia ( Regierung von Unterfranken; 621-2531 . 01-2/05 and 55 . 2-2531 . 01-73/07 ) . All experiments were performed with the natural E . multilocularis isolate JAVA [32] which was propagated in Mongolian jirds ( Meriones unguiculatus ) as described [33] . Isolation of metacestode tissue and axenic cultivation of metacestode vesicles was performed essentially as described previously by Spiliotis and Brehm [33] . For the isolation of protoscoleces , parasite tissue was isolated from infected jirds and homogenized as described [34] . The homogenate was subsequently filtered through a nylon mesh of 150 µm pore size , thus separating protoscoleces from larger pieces of metacestode tissue . The flow through was subsequently filtered through a nylon mesh of 30 µm pore size , separating protoscoleces from single cells and small cell clumps . Protoscoleces were then washed off the nylon mesh with sterile PBS and separated from equally sized metacestode vesicles by microscope-aided , manual picking with a pipette tip prior to applying axenic cultivation conditions in order to eliminate eventual host remnants [33] For the isolation of primary cells , axenically cultivated metacestode vesicles were mechanically sheared and trypsin-digested essentially as previously described [28] . Primary cells were then directly cultivated in hepatocyte-conditioned medium supplemented with reducing agents under a nitrogen atmosphere . After one week of cultivation under axenic conditions [33] , [35] , the different larval stages ( primary cells , metacestode vesicles , protoscoleces ) were analyzed for host cell contamination by organism-specific PCR . Chromosomal DNA was isolated from the larvae and from liver tissue of a non-infected jird using the DNeasy isolation kit ( Qiagen ) . Part of the parasite specific gene elp ( ezrin-radixin-moesin-like [36] ) was amplified using the primers Em10-15 ( 5′-TCC TTA CCT TGC AGT TTT GT -3′ ) and Em10-16 ( 5′-TTG CTG GTA ATC AGT CGA TC-3′ ) . As a control for host-DNA contamination , a previously described β-tubulin-encoding gene from Meriones unguiculatus was used [37] , employing primers Tub12-UP and TUB12-ST as described [38] . In vitro cultivated material of all three larval stages was isolated and cell lysates were generated by first passing larval material repeatedly through a pipette tip , followed by one washing step in 1×PBS , and subsequent treatment with 50 µl of 2× STOP mix ( 2 ml 0 . 5 M Tris–HCl pH 6 . 8 , 1 . 6 ml glycerol , 1 . 6 ml 20% SDS , 1 . 4 ml H2O , 0 . 4 ml 0 . 05% ( w/v ) bromphenol blue , 7 µl β-mercapto-ethanol per 100 µl ) and boiling for 10 min at 100°C . Proteins were separated by SDS-PAGE and analyzed by Western blotting using an antibody directed against β-actin ( Cell signalling technology®; No . 4967 ) of a wide variety of metazoan organisms . Images were subsequently analyzed for the relative expression of β-actin using the Image-J program ( http://rsb . info . nih . gov/ij/ ) [39] . The relative expression transcribed as values of area under the curve ( AUC ) was used to normalize the β-actin content of each sample . In a first set of experiments , different amounts of in vitro cultivated primary cells , metacestode vesicles and protoscoleces were analyzed ( Figure S1A ) and the relative β-actin content of each sample was then used as a basis for normalization . Based on previous studies showing that , in vivo , oncosphere-derived stem cells develop into mature metacestode vesicles within 2–3 weeks upon infection [3]–[6] , we first determined the amount of primary cells which , in our in vitro system , lead to the production of metacestode vesicles within the same time . This was the case when we used 1/6th of the amount of primary cells that can be isolated from 40 ml of metacestode vesicles ( Figure S1B ) . This amount was defined as 1 Unit and contained ∼600 µg of total protein ( Figure S1B ) . We then carried out calculations for the remaining larval stages and found that 2000 protoscoleces as well as 4 metacestode vesicles ( 2 months of age ) of a diameter of 5 mm after 1 week of axenic cultivation represented 1 Unit and also contained ∼600 µg of total protein ( Figure S1B ) . The reliability of this quantitative approach was further assessed by comparing one half unit of parasite material from each stage ( i . e . 1000 protoscoleces , 2 metacestode vesicles of a diameter of 5 mm and 1/12th of the amount of primary cells that can be isolated from 40 ml metacestode vesicles ) which , as shown in Figure S1C , also resulted in comparable β-actin content . C57BL/6 mice were purchased from Charles River/Wiga ( Sulzfeld , Germany ) and TCR transgenic OT2 B6 mice were a kind gift of Prof . F . Carbone ( Melbourne , Australia ) . All mice were bred within the animal facility of the Institute of Virology and Immunobiology , University of Würzburg , under specific pathogen-free conditions . Female mice were used at the age of 6–14 weeks . Single cell suspensions were obtained from the spleen of C57BL/6 mice by mechanically squeezing the tissue with glass slides in cold PBS and filtered through a 70 µm nylon cell strainer . Red blood cells in the filtrate were lysed with 1 , 4% NH4Cl for 5 minutes at 37°C , and the splenocytes were washed in R10 medium , that is RPMI 1640 ( GIBCO BRL ) supplemented with penicillin ( 100 U/ml , Sigma , Deisenhofen , Germany ) , streptomycin ( 100 µg/ml , Sigma ) , L-glutamine ( 2 mM , Sigma ) , 2-mercaptoethanol ( 50 µM , Sigma and 10% heat-inactivated fetal calf serum ( FCS , PAA Laboratories , Parsching , Austria ) . Cell counts were subsequently determined using the trypan blue ( No . 26323 , Biochrom , Berlin , Germany ) exclusion test on a bright-lined Neubauer counting chamber . DC were generated from the bone marrow ( BM ) precursors of C57BL/6 mice as previously described [40] . Briefly , BM precursor cells were cultured for 8 days with GM-CSF . At day 8 , non-adherent DC ( 70–80% CD11c+ cells ) were harvested and seeded at a density of 106 cells/ml R10 culture medium . A comparable amount of axenically cultivated parasite material , normalized for β-actin content , from each of the three larval stages was used throughout the stimulation process . Tissue culture inserts ( Greiner Bio-One ) of 1 µm pore size with or without larvae were thoroughly washed in R10 medium to completely remove hepatocyte-conditioned medium , were then added to DC or splenocyte cultures , and kept at 37°C in the presence of 5% CO2 for different time points . For LPS stimulation experiments , inserts containing parasite material were removed after 24 h , DC were harvested and re-plated at an equal number of living cells ( 5×105 cells/ml ) in R10 culture medium with or without 0 . 1 µg/ml lipopolysaccharide ( LPS; E . coli 0127:B8; Sigma Aldrich ) for additional 48 h . Upon completion , DC or splenocyte viability was determined by trypan blue exclusion ( No . 26323 , Biochrom , Berlin , Germany ) on a bright-lined Neubauer counting chamber . Flow cytometric assessment of DC surface staining was then performed using fluorochrome-conjugated antibodies ( anti-mouse ) against the surface lineage marker CD11c ( CD11c-PE-Cy5 . 5 , Caltag Laboratories ) , MHC II ( eBiosciences ) and CD86 ( B7-2-FITC , eBiosciences or B7-2-PE , BD Pharmingen ) . Splenocytes were stained for CD19 ( CD19-pecy5 , BD Pharmingen ) as a specific marker for B cells and an exclusion marker for T cells within lymphocytes . To monitor the level of DC apoptosis , annexin-V binding buffer ( BD Pharmingen ) and FITC-conjugated annexin-V ready-to-use solution ( BD Pharmingen ) were used coupled to 7-AAD staining solution ( BD Pharmingen ) . To assess DC maturation , marker-specific antibodies ( CD11c , MHC II and CD86 ) were applied and after 30 min incubation at 4–8°C in the dark , cells were washed twice with FACS buffer ( 3% FCS , 0 . 1% NaN3 in PBS ) and fixed with 1% ( v/v ) formaldehyde in PBS . The staining procedure was identically conducted for CD 19 on splenocytes . In DC apoptosis assays , stimulated DC along with UV-irradiated DC ( positive control for apoptosis ) at a peak intensity of 9000 mW/cm2 at the filter surface and a peak emission of 313 nm ( trans-illuminator ) , were directly resuspended in 50 µl of 1× annexin-V binding buffer . Next , 5 µl of 7-AAD and 2 µl of annexin-V–FITC were added to the tubes and incubated for 15 minutes at RT . Cells were then resuspended in 200 µl of 1× annexin-V binding buffer then acquired on a FACSCalibur™ ( Beckton Dickinson ) cytometer , equipped with CellQuest software . Results were further analyzed with FlowJo software ( Tree Star , USA ) . After stimulation of DC , the production of IL-6 , IL-10 , and IL-12p70 was measured in supernatant using sandwich enzyme-linked immunosorbent assays ( ELISA , OptEIA kits , BD Pharmingen ) according to the manufacturer's instructions . The kits detection limits were of 39 pg/ml for IL-12p70 and 19 pg/ml for IL-10 and IL-6 . Spleens and lymph nodes from 6–14 weeks old OT2 mice were isolated , and the separated splenocytes and lymph nodes cells , obtained as described above ( Splenocytes isolation ) , were resuspended in cold PBS . CD4+ cells were isolated using an EasyStep negative selection mouse CD4+ T cell enrichment kit ( Stemcell Technologies ) . After separation , the purity of T-cell preparation was routinely higher than 90% , as determined by flow cytometry . CD4+ T cells were subsequently enriched for CD25− cells using Miltenyi Biotec's LD columns with a suitable MACS separator usually achieving 90–95% of purity . DC were incubated with 3-fold higher numbers of OT2 CD4+CD25− T cells and 200 ng/ml of OVA protein ( Sigma , grade V ) supplemented or not with parasite larvae E/S-products ( supernatant of equal amounts of larvae kept in medium for 7 days ) . After 5 days of co-culture , the cells were harvested and stained using the T-reg detection Kit ( Miltenyi Biotec ) prior to flow cytometric analysis . All results were expressed as mean ± standard deviation ( SD ) . Differences observed between groups were evaluated using the Wilcoxon/Mann-Whitney U test , a nonparametric test that does not assume normality of the measurements ( it compares medians instead of means ) . Values of p<0 . 05 were considered statistically significant . All statistical analysis were performed with STATISTICA version 8 . 0 . 725 . 0 ( StatSoft GmbH )
The morphology of the three different E . multilocularis larval stages investigated in this study is depicted in Figure 1A . Primary cells were isolated from the GL of metacestode vesicles and seeded into culture dishes where they formed aggregates with central cavities within one week of cultivation . As previously outlined , these primary cell aggregates closely resemble the early developing metacestode both morphologically and physiologically , and routinely result in the production of fully mature metacestode vesicles after 3–4 weeks of cultivation [28] . Furthermore , primary cell aggregates express factors such as members of the EG95/W45 protein family ( or host protective oncospheral antigens ) that are specifically present in taeniid oncospheres and are known to play an important role in early parasite establishment [30] . Primary cell aggregates thus closely mimic the E . multilocularis larval stage at the onset of the oncosphere-metacestode metamorphosis [29] , [30] . In all experiments , primary cell cultures were carefully checked for the absence of mature , LL-containing metacestode vesicles ( Figure 1Aa ) . Mature metacestode vesicles had a size of approximately 5 mm ( diameter ) and were completely equipped with a LL surrounding a cellular GL ( Figure 1Ab ) . Protoscoleces ( 50–100 µm in size ) were covered by a tegument and were used in a non-activated , dormant state ( i . e . no pre-activation with low pH and trypsin ) , as they typically occur within metacestode vesicles in the intermediate host ( Figure 1Ac ) . After one week cultivation under axenic conditions [33] , [35] , the absence of contaminating , cellular host material in all parasite samples was confirmed by organism-specific PCR ( Figure 1B ) . Parasite material of each of the three larval stages was subsequently normalized on the basis of β-actin content ( Figure 1C; Figure S1 ) . In all subsequent DC co-cultivation procedures , comparable amounts of parasite material were used with 1 Unit defined as 2000 protoscoleces , 4 metacestode vesicles of 5 mm of diameter , and 1/6th of primary cells generated from 40 ml of metacestode vesicles after 1 week of in vitro culture ( Figure 1C ) . In a first set of experiments , the influence of E . multilocularis E/S-products on DC viability was tested . To this end , E . multilocularis larval material ( 1 Unit each for all three larval stages ) was co-incubated for 48 h with DC , physically separated through a transwell system ( 1 µm pore size ) , and the number of viable DC was assessed by trypan blue exclusion . As shown in Figure 2A , co-incubation of DC with primary cells and metacestode vesicles led to greatly reduced viability ( 30–40% surviving cells compared to the control ) , whereas a still significant , but reduced killing effect was observed in the presence of protoscoleces ( 76 , 6+/−3 , 7% survival ) . To exclude the possibility that cell death in DC-parasite co-cultures merely resulted from starvation due to the presence of proliferating larvae , conditioned medium of a comparable amount of primary cells was tested on DC . To this end , supernatant from primary cell cultures ( 7 days old ) was collected , sterile filtered , and added to fresh cultures of DC . As depicted in Figure 2B , even in the absence of proliferating larvae , conditioned medium similarly induced DC death . In order to assess whether Echinococcus E/S-products do have general cytolytic effects that might account for the observed DC death , a hemolysis assay was carried out in which parasite material was co-incubated with human erythrocytes . However , as shown in Figure S2 , no such effects were observed for any of the larval stages . We further tested whether E . multilocularis metacestode vesicles , which displayed the strongest killing effect on DC , could also affect other immune effector cells . Splenocytes from C57BL/6 mice were exposed to metacestode vesicles through a transwell system for 48 to 72 h and host cell viability was assessed by trypan blue exclusion . As shown in Figure 2C , in contrast to what we observed for BMDC , splenocyte viability was not affected by E/S-products of metacestode vesicles . We also specifically analyzed the CD19+ ( B cells ) and CD19− ( primarily T cells ) splenic lymphocytes for the effect of MCE/S on viability . As observed for the whole splenocyte population , the splenic B and T cell proportions were not altered upon 48–72 h of exposure to MCE/S ( Figure 2D ) . Taken together , these data indicated that E/S-products of E . multilocularis larvae , particularly those that are released by primary cells and metacestode vesicles , induce murine DC death , but do not have general cytolytic properties and do not lead to killing in whole spleen cell preparations or alter the splenic B and T cell compartments in vitro . To examine by which mechanism the E . multilocularis E/S-products induced DC death , Annexin-V/7-AAD dual staining was performed to differentiate necrotic ( 7AAD+ ) from apoptotic ( Annexin-V+ ) cells . DC were separately exposed to comparable amounts of parasite material from each of the larval stages through transwells for 24 h , harvested and processed for staining . As shown in Figure 2E , F , following exposure of DC to E/S-products of primary cells or metacestode vesicles , 2-fold more DC underwent apoptosis than DC exposed to protoscolex E/S-products , which showed a similar rate of apoptosis as untreated DC . Taken together , these results indicated apoptosis as the primary mechanism by which E/S-products from primary cells and metacestode vesicles induce DC death . Having shown that E/S-products of E . multilocularis primary cells and metacestode vesicles induce apoptosis in part of the co-incubated DC , we subsequently analyzed the fate of the surviving DC concerning maturation and cytokine release . To this end , DC were first incubated for 72 h with each of the three different E . multilocularis larval stages , separated by transwells . Subsequently , DC maturation was assessed by measuring the expression of surface markers MHCII and CD86 by flow cytometry ( Figure 3A , B ) . Interestingly , E/S-products from primary cells and metacestode vesicles significantly inhibited the spontaneous DC maturation rate as compared to untreated DC , which was not observed for E/S-products of protoscoleces ( Figure 3A , B ) . To investigate whether parasite larvae also alter DC cytokine release , DC were first exposed to each of the parasite stages through transwells for 24 h after which the parasite larvae were removed . Upon this brief exposure to the various larvae , the DC were harvested , counted and an equal number of surviving DC were re-seeded in fresh medium for an additional 48 h . Supernatant was then harvested and the level of secreted IL-12 , IL-6 and IL-10 was assessed by ELISA . As expected , bacterial LPS , used here as a positive control , led to a strong induction of all three cytokines in DC ( Figure 3C ) . On the other hand , none of the parasite E/S-products was able to trigger IL-12 release . However , substantial amounts of IL-10 were produced upon exposure of DC to E/S-products of primary cells , whereas those of protoscoleces induced the production of IL-6 . Following challenge by metacestode E/S-products , neither IL-6 nor IL-10 production were induced in DC ( Figure 3C ) . Taken together , these results indicated that E/S-products of primary cells and metacestode vesicles inhibited the ability of DC to spontaneously mature in vitro , and failed to promote ( or blocked ) the release of detectable amounts of the pro-inflammatory and Th1-associated cytokine , IL-12 . Interestingly E/S-products of primary cells additionally acted on DC to induce IL-10 secretion , a feature which was not seen with E/S-products of the metacestode . In contrast , E/S-products of protoscoleces fostered DC maturation and induced IL-6 but not IL-12 secretion by DC . The consistent absence of detectable amounts of IL-12 in culture supernatant of DC after treatment with E . multilocularis larvae suggested that parasite E/S-products either failed to induce the secretion of this Th1-associated cytokine , or simply impaired its production by treated DC . Since there is increasing evidence that parasite survival within the intermediate host depends on the ability to deviate immune polarization away from a potential parasitocidic Th1 type [8] , we examined whether exposure of DC to E . multilocularis E/S-products could have an influence on subsequent DC maturation by LPS , a stimulus usually associated with strong IL-12 release . To this end , DC were first incubated with each of the larval stages physically separated through transwells for 24 h . After this brief incubation time , the parasite larvae were removed and the DC were harvested , counted and an equal number of surviving DC were re-seeded in fresh medium containing 0 . 1 µg/ml of LPS for an additional 48 h . At the end of this second incubation period , DC were assessed for their level of maturation ( MHCII and CD86 surface molecule expression ) and DC supernatant was analyzed by ELISA for the presence of IL12p70 , IL10 and IL6 . As shown in Figure 4A/B , in contrast to the situation for protoscolex co-cultures , LPS did not induce maturation when DC had previously been cultivated in the presence of primary cells or metacestode vesicles . Furthermore , pre-exposure to E/S-products of all three larval stages significantly altered the cytokine profile of DC upon subsequent LPS stimulation , again leading to a strongly diminished release of the Th1 associated cytokine IL-12 ( Figure 4C ) . Taken together , our results indicate that exposure of DC to E/S-products of primary cells inhibits maturation in response to LPS , diminishes the ability to produce IL-12 and IL-10 , but induces IL-6 production . E/S-products of metacestode vesicles act similarly on DC , but have a reduced effect on LPS-induced IL-10 and IL-6 production . In contrast , E/S-products of protoscoleces do not inhibit the expression of MHCII and CD86 surface markers by LPS-stimulated DC , but strongly affect LPS-induced IL-12 production . A recent report has suggested an implication of Foxp3+ regulatory T cells ( T-reg ) in E . multilocularis larval establishment and/or persistence within the intermediate host [15] . Furthermore , a key role of DC in inducing the de novo generation of Foxp3+ T-reg is well established [41]–[43] . Using an in vitro OVA peptide-based assay for the co-cultivation of DC and CD4+ T cells [44] we therefore tested whether additionally present E/S-products of E . multilocularis larvae could lead to an expansion of CD4+CD25+Foxp3+ T cells , commonly assumed to be T-reg [41]–[43] . To this end , co-cultures of freshly generated DC ( day 8 ) and naïve CD4+CD25− T cells from OT2 αβ TCR-transgenic mice at a DC/T cell ratio of 1/3 , supplemented with OVA peptide , were exposed to E/S-products of all three larval stages . Upon 5 days of incubation , the cells were harvested and stained for Foxp3+ T-reg-specific cell markers ( CD4 , CD25 or IL2Rα chain and Foxp3 ) . Interestingly , in contrast to co-cultures of DC and T-cells with E/S-products of primary cells and protoscoleces , there was a significant expansion ( 2 . 5-fold ) of the population of T-reg ( CD4+CD25+Foxp3+ ) in co-cultures that included E/S-products of metacestode vesicles ( Figure 5 ) . These results indicated that at least the developmental stage that characterizes the chronic phase of AE , the metacestode , is able to induce de novo CD4+CD25+Foxp3+ T-reg conversion in vitro in the presence of DC .
As typical in the case of helminth infections , AE is a long-lasting and chronic disease that is most probably associated with parasite-induced , immunosuppressory mechanisms around the primary site of infection [8] . For a number of nematode and trematode systems , research during recent years has demonstrated a crucial role of T-reg in the respective immunosuppressive mechanisms and emphasized the importance of DC in the induction of helminth-associated Th2- and tolerogenic immune responses [11] , [12] . Compared to nematode and trematode infections , immunomodulatory functions of DC in cestode infections have drawn significantly less attention , although this is clearly an emerging field since several studies concerning the influence of parasite products on DC maturation and cytokine secretion profiles have been conducted very recently . In two of these recent reports , Reyes et al . [45] and Terrazas et al . [46] investigated the effects of E/S-products of Taenia crassiceps cysticerci , representing the metacestode larval stage of this Taenia infection model , on the activation of murine DC . These authors observed impaired DC maturation in response to TLR dependent stimuli , particularly when DC of infection susceptible mouse strains were pre-incubated with parasite E/S-products [45] . In the case of E . granulosus , a species closely related to E . multilocularis , the effects of hydatid cyst fluid ( HCF ) and isolated antigen B ( AgB ) , a major constituent of HCF , were tested and led to DC maturation as well as DC cytokine profiles that were indicative of Th2 immune responses [19] , [20] . However , whether these interactions are of major relevance in vivo remains questionable since intact parasite tissue usually prevents direct contact between HCF and host immune effector cells , and the spectrum of metacestode E/S-products does not necessarily overlap with the spectrum of proteins present in HCF . Although it is generally assumed that AgB might leak out of intact metacestode vesicles or be released early during an infection from damaged metacestode material [47] , we could not detect AgB in the E/S-products of in vitro cultivated E . multilocularis metacestode vesicles despite the fact that this component was well expressed in HCF [48] . Crude metacestode antigen preparations containing vesicle fluid , somatic parasite proteins and contaminating host components [16] as well as isolated vesicle fluid of E . multilocularis [20] were also already tested concerning their effects on DC and failed to induce maturation as did a purified mucin-type glycoprotein ( Em2 ) that is usually expressed at the surface of LL-containing metacestode vesicles [49] , [50] . Hence , although all these reports indicate that larval cestode parasite products can exert immunomodulatory effects on host DC , depending on the source and form of application , their precise nature and role during the course of an infection of the intermediate host remains elusive so far . What became clear through a recent in vivo investigation on experimentally infected mice , on the other hand , was that , at least in the chronic phase of experimental AE , peritoneal DC display a significant down-regulation of surface markers that are associated with DC maturation , and over-expressed TGF-β mRNA , which might lead to an induction of T-reg in this phase of the disease [15] , [51] . A potential role of T-reg in human AE has also already been suggested [8] based on the fact that immune-suppressive TGF-β and IL-10 , major cytokines that are released by T-reg [52] , can be predominantly found in the immediate vicinity of actively proliferating parasite tissue . In order to more closely mimic the situation at the site of infection , we utilized in this study a cultivation system by which actively secreted E/S-products of living parasite material can be tested on DC . Furthermore , in addition to parasite components that are produced during late stages of the infection ( metacestode and eventually protoscoleces ) , we also included larval material that represents early stages of the infection , prior to the establishment of metacestode vesicles , in which E . multilocularis should be highly susceptible to host attacks due to the absence of a protective LL [7] . First , although co-incubation with protoscolex E/S-products clearly induced DC maturation , no activation could be observed upon co-incubation with E/S-products of primary cells and metacestode vesicles . The cells were not only affected in the expression of surface activation markers but also failed to secrete pro-inflammatory cytokines . This inhibitory effect was even apparent in the presence of strong stimuli of TLR signaling since DC pre-incubated with E/S-products of primary cells and metacestode vesicles did not mature in response to LPS , whereas pre-incubation with E/S-products of protoscoleces had no such effect . Furthermore , in a much more pronounced manner than protoscolex compounds , E/S-products of both primary cells and metacestode vesicles induced DC death mediated by apoptosis . To our knowledge , this is the first report on the induction of apoptosis in host DC in response to cestode larval material , which should have important implications concerning immuno-suppressive activities , particularly at the very early stage of the infection . On the one hand , the induction of apoptosis in DC should be beneficial to the parasite since it depletes immune effector cells around the early parasite lesions that are important to induce inflammatory immune responses . Moreover , it is well established that apoptosis , extrinsically triggered by infectious agents such as viruses , parasites , or bacteria , usually results in a bystander effect of induced immunosuppression [53] . In parasitic helminths , the induction of DC apoptosis has already been reported for microfilariae of the nematode Brugia malayi which , as also shown herein for E . multilocularis , strongly limited their capacity to produce pro-inflammatory IL-12 , and prevented T cell activation and proliferation [54] . Previous in vitro studies further demonstrated that apoptotic DC are rapidly taken up by immature DC , which prevents subsequent maturation of immature DC in response to TLR stimuli [53] . It is , therefore , conceivable that the strongly diminished ability of DC that were pre-incubated with E/S-products of primary cells and the metacestode to LPS , as observed in our study , is indirectly mediated by the induction of apoptosis in a subset of immature DC , rather than by direct inhibition of DC maturation through parasite E/S-products . Since the uptake of apoptotic DC induces immature DC to secrete TGF-β , which induces differentiation of naïve T cells into Foxp3+ T-reg [53] , E/S-products of the metacestode , and particularly of primary cells , could thus establish a strongly immunosuppressive environment around parasite lesions already at the beginning of an infection . As in the case of E/S-products produced by B . malayi microfilariae [54] , we can currently only speculate about the molecular nature of Echinococcus E/S factors that might induce DC apoptosis . Among the various host-derived compounds that can extrinsically trigger DC apoptosis are ligands of the tumor necrosis factor ( TNF ) superfamily as well as glucocorticoids [53] and B . malayi microfilariae have already been demonstrated to induce DC apoptosis by triggering TNFα-dependent signaling mechanisms [55] . Although no TNF-like ligand has been described so far in Echinococcus or any other flatworm , there has been a recent report on the presence of a TNFα-receptor like surface protein in Schistosoma mansoni which presumably interacts with host TNFα [56] . Our own preliminary analyses on the E . multilocularis genome , which is currently being sequenced [29] , [30] , revealed that a very similar receptor is also expressed by cestodes ( data not shown ) . Bioinformatically , TNF-ligands are difficult to identify in raw sequence data , which might be the reason why so far no such molecule was identified in the Schistosoma or Echinococcus genomes . However , the presence of a respective receptor in these organisms implies that they might also express cognate ligands , which subsequently could bind to members of the TNF-receptor family on host cells ( such as DC ) , thus triggering apoptosis . Apart from components of the TNF signaling machinery , it has recently also been demonstrated that cestode larvae ( T . crassiceps cysticerci ) are capable of producing steroid hormones [57] . Although in this system only the production of sex steroids has been tested , it is conceivable that they also produce glucocorticoids which might , either together with TNF-ligands or as an alternative , be involved in triggering host DC apoptosis . By utilization of the culture system established in this work , these alternatives can now be addressed . A marked difference between DC that were incubated in the presence of E/S-products of metacestode vesicles and primary cells was that , in the latter case , the production of anti-inflammatory IL-10 was significantly induced . To our knowledge , an elevated expression of IL-10 coinciding with DC apoptosis has so far never been described , indicating that this effect was not provoked by the elevated induction of DC apoptosis through E/S-products of primary cells , when compared to those of the metacestode . Hence , we rather suggest that primary cells secrete a set of factors that differs from E/S-products of the metacestode and contains additional components that are able to induce the expression of IL-10 by non-activated DC . This hypothesis is supported by the differential influence of E/S-products from primary cells and metacestode vesicles that we have observed in T-reg conversion assays . Only E/S-products of metacestode vesicles , but not those of primary cells ( or protoscoleces ) , were able to significantly increase the number of CD4+CD25+Foxp3+ regulatory T cells in vitro . Although we cannot presently tell whether the in vitro T-reg conversion was exclusively mediated by the modified DC , or whether there is also a direct influence of parasite E/S-products on CD4+ T cells , these data nevertheless clearly support an emerging picture that points at Foxp3+ T-reg cells as potential mediators of the fine tuning of the host immune system during metacestode establishment and growth within the intermediate host [15] . Furthermore , our data suggest that an expansion of Foxp3 expressing T cells during chronic ( peritoneal ) AE , as observed by Mejri et al . [15] , might not simply be an intrinsic consequence of an ongoing immune response , but that the parasite actively induces Foxp3+ T-reg through its E/S-products . The factor ( s ) and mechanism ( s ) involved in the modulation of DC maturation and function are currently subject to ongoing investigations . These include studies on the possibility that Echinococcus E/S-products might directly interact with TLR ligands rendering the latter less able to elicit a ‘normal’ response from DC , or may be acting as antagonists of TLR-ligand binding interactions [11] . Furthermore , parasite E/S-products may affect DC directly through interactions with non-TLR pattern recognitions receptors such as DC-SIGN , Dectin family members [11] . In previous studies on other helminth systems , secreted compounds such as filarial cystatins [58] were shown to induce the expression of IL-10 in antigen presenting cells , including non-activated DC . Interestingly , our own preliminary analyses of the E . multilocularis genome sequence indicate that related molecules are also encoded by the cestode , although further experimentation is clearly necessary concerning a possible secretion of these molecules by PC or whether they exert immunomodulatory activities comparable to those of filarial cystatins . Regarding T-reg conversion , the so far best characterized component that elicited similar in vitro effects as E/S-products from metacestode vesicles was a secreted compound of the nematode Heligmosomoides polygirus with TGF-β-like activities [59] . Due to the fact that TGF-β-signaling mechanisms have already evolved very early in animal evolution , TGF-β-like cytokines are expressed by a wide variety of free-living , but also parasitic invertebrates [60] , [61] . Notably , at least one gene that encodes a structural homolog of mammalian TGF-β is also present on the genome of E . multilocularis [62] , and the involvement of this component in the in vitro T-reg conversion process induced by metacestode E/S-products is currently investigated by us using the in vitro cultivation models established in this study . In sharp contrast to co-incubation with primary cells and metacestode vesicles , DC exposed to E/S-products of protoscoleces were clearly activated , as assessed by up-regulation of surface activation markers ( MHCII and CD86 ) , secreted elevated levels of IL-6 ( but no IL-10 ) , and strongly impaired the ability of DC to produce IL-12 in response to TLR stimuli ( LPS ) . This phenotype resembles that of DC that had been incubated in the presence of E . granulosus HCF and isolated AgB [19] , [20] . However , in contrast to these investigations , DC incubated with protoscolex compounds in our study did not release elevated levels of IL-10 , as reported by Rigano et al . [20] , or IL-12 , as reported by Kanan and Chain [19] . This is most probably due to the fact that the spectrum of E/S-products of protoscoleces does not fully overlap with the content of HCF since , for example , AgB is only weakly expressed by protoscoleces [30] , [63] . In general , however , the phenotype of DC upon co-incubation with E/S-products of protoscoleces in our study is largely comparable to that of DC incubated with certain Trypanosoma antigens which have been closely associated with the induction of Th2-dominated immune responses [64] . Whether the Th2 immune response that is characteristic of the chronic stage of AE [8] is provoked ( or supported ) by direct contact between protoscoleces and DC within the intermediate host remains highly questionable , since this larval stage is only produced very late in the infection and direct contact between protoscoleces and host cells is usually prevented by the parasite's surface layers . Furthermore , Th2-dominated immune responses can also be observed in chronic AE under conditions in which no protoscoleces are produced [65] . However , since intestinal luminal infections by adult cestodes are associated with Th2 immune responses [66] , the phenotype we observed in this study for DC exposed to E/S-products of protoscoleces could rather be associated with immunological processes that are relevant for an infection of the definitive host [66] . In any case , the marked differences between the responses of DC to E/S-products of early versus late developmental stages of E . multilocularis clearly demonstrates that an induction of tolerance in DC is not a general characteristic of Echinococcus material , but rather that the E/S repertoire of primary cells and metacestodes has specifically evolved to fulfill these purposes . Care should therefore be taken in the interpretation of results that have been obtained concerning the immune response during echinococcosis ( intermediate host infection ) by using co-incubation-systems of Echinococcus protoscoleces with host cells [21]–[27] or by employing the mouse model of peritoneal , protoscolex-induced secondary alveolar echinococcosis for short-term infections [67] . In conclusion , in this study we provide for the first time evidence for the induction of apoptosis in host DC through E/S-products of early infectious stages of E . multilocularis . We further show that primary cells , as representative of the oncosphere stage that undergoes metamorphosis towards the metacestode , are able to induce poorly responsive , IL-10 secreting DC in vitro . This effect is somewhat reduced at the chronic stage ( metacestode ) , leading to poorly responsive , immature DC , but a Foxp3+-T-reg-inducing environment , and is no longer present in the protoscolex stage ( table 1 ) . Although our study concentrated on in vitro interactions between parasite larvae and DC , thus excluding the possible influence of other immune effectors or epithelial cells , the clear induction of poorly responsive , apoptotic and IL-10 secreting DC in response to primary cells suggests that a similar mechanism might also be operative in the tissue surrounding the early metamorphosing oncosphere . If so , this process might be important for an early establishment of the parasite during a phase of relatively high vulnerability to the host immune system , whereas in the chronic phase , after production of the LL , a slightly altered profile of E/S-products that mainly induces T-reg could support long-term persistence and infiltrative growth of the metacestode , as previously suggested [15] . The molecular nature of Echinococcus E/S-products that are responsible for these effects is currently being investigated by us using the available genome sequence information [29] , [30] , recently established methods for genetic manipulation of primary cells [68] , and the cultivation settings established in this work . | Parasitic helminths are inducers of chronic diseases and have evolved mechanisms to suppress the host immune response . Mostly from studies on roundworms , a picture is currently emerging that helminths secrete factors ( E/S-products ) that directly act on sentinels of the immune system , dendritic cells , in order to achieve an expansion of immunosuppressive , regulatory T cells ( T-reg ) . Parasitic helminths are currently also intensely studied as therapeutic agents against autoimmune diseases and allergies , which is directly linked to their immunosuppressive activities . The immunomodulatory products of parasitic helminths are therefore of high interest for understanding immunopathology during infections and for the treatment of allergies . The present work was conducted on larvae of the tapeworm E . multilocularis , which grow like a tumor into surrounding host tissue and thus cause the lethal disease alveolar echinococcosis . The authors found that E/S-products from early infective larvae are strong inducers of tolerogenic DC in vitro and show that E/S-products of larvae of the chronic stage lead to an in vitro expansion of Foxp3+ T cells , suggesting that both the expansion of these T cells and poorly responsive DC are important for the establishment and persistence of E . multilocularis larvae within the host . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"immune",
"cells",
"clinical",
"immunology",
"immunity",
"immunology",
"biology",
"microbiology",
"zoology"
] | 2012 | Excretory/Secretory-Products of Echinococcus multilocularis Larvae Induce Apoptosis and Tolerogenic Properties in Dendritic Cells In Vitro |
Two highly similar RNA polymerase sigma subunits , σF and σG , govern the early and late phases of forespore-specific gene expression during spore differentiation in Bacillus subtilis . σF drives synthesis of σG but the latter only becomes active once engulfment of the forespore by the mother cell is completed , its levels rising quickly due to a positive feedback loop . The mechanisms that prevent premature or ectopic activation of σG while discriminating between σF and σG in the forespore are not fully comprehended . Here , we report that the substitution of an asparagine by a glutamic acid at position 45 of σG ( N45E ) strongly reduced binding by a previously characterized anti-sigma factor , CsfB ( also known as Gin ) , in vitro , and increased the activity of σG in vivo . The N45E mutation caused the appearance of a sub-population of pre-divisional cells with strong activity of σG . CsfB is normally produced in the forespore , under σF control , but sigGN45E mutant cells also expressed csfB and did so in a σG-dependent manner , autonomously from σF . Thus , a negative feedback loop involving CsfB counteracts the positive feedback loop resulting from ectopic σG activity . N45 is invariant in the homologous position of σG orthologues , whereas its functional equivalent in σF proteins , E39 , is highly conserved . While CsfB does not bind to wild-type σF , a E39N substitution in σF resulted in efficient binding of CsfB to σF . Moreover , under certain conditions , the E39N alteration strongly restrains the activity of σF in vivo , in a csfB-dependent manner , and the efficiency of sporulation . Therefore , a single amino residue , N45/E39 , is sufficient for the ability of CsfB to discriminate between the two forespore-specific sigma factors in B . subtilis .
When cells of Bacillus subtilis enter stationary phase and face severe nutrient depletion , they may embark into a developmental pathway that results in the production of a dormant , highly resistant endospore [1] . Sporulation involves the asymmetric division of the rod-shape cell into a smaller forespore , the future spore , and a larger mother cell . Soon after asymmetric cell division , the mother cell engulfs the forespore , eventually releasing it as a free protoplast within its cytoplasm . Following engulfment completion , the forespore becomes encased in a series of protective layers after which it is released into the environment through lysis of the mother cell [1] . Underlying the differentiation process are mother cell- and forespore-specific programs of gene expression administered by a cascade of cell type-specific RNA polymerase sigma factors . σF and σE govern the initial stages in development in the forespore and in the mother cell , respectively . At late stages of development , σF is replaced by σG ( Figure 1A ) , whereas σK replaces σE . The sporulation-specific sigma factors are produced prior to their period of activity , and maintained inactive until the successful conclusion of key morphological events during development . Both σF and σE are synthesized in the predivisional cell . Proper septation is a prerequisite for the activation of σF in the forespore and soon after a signaling pathway initiated by σF leads to the activation of σE in the mother cell . Likewise , synthesis of σG and σK is initially driven by σF and σE , respectively . However , σE-dependent gene expression is required for the activation of σG following engulfment completion and when active , σG initiates a signaling pathway that causes the activation of σK ( [1]–[3] see also below ) . The double responsiveness of the cell-type specific σ factors to proper morphogenesis and to intercompartmental signaling pathways effectively links the forespore and mother cell programs of gene expression and keeps gene expression in close register with the course of morphogenesis . Importantly , proper timing of sigma factor activation is essential for the fidelity of the developmental process [reviewed by [1]–[3]] . This study addresses the mechanisms involved in the regulation of the activity of σG . Productive transcription of the sigG gene ( coding for σG ) is controlled by σF [4] , [5] . However , sigG is not transcribed as soon as σF becomes active . The delay appears to result from an as yet poorly understood dependency of sigG transcription upon the activity of σE in the mother cell [6] , [7] . σG can be detected in the forespore towards the end of the engulfment sequence , but its window of activity begins only after engulfment completion . Activity of σG requires the assembly of a novel type of secretion system formed by eight mother cell proteins ( AA through AH ) coded for by the σE-controlled spoIIIA operon , and by the forespore-specific , σF-controlled protein SpoIIQ [8]–[14] , with the assistance of the membrane protein translocase SpoIIIJ [8] , [15]–[18] . The SpoIIIA-SpoIIQ complex spans the intermembrane space that separates the forespore and the mother cell establishing a direct connection between the cytoplasm of the two cells [8] , [10] , [14] , [19] . Recent work has lead to the concept that the channel acts as a feeding tube , maintaining the potential for macromolecular synthesis when the forespore becomes isolated from the external medium [9] . This model brings the important implication that the activation of σG in engulfed forespores does not necessarily involve counteracting a specific inhibitor or inhibitors of σG . However , once active , σG recognizes its own promoter , creating a positive feedback loop that causes its levels to increase rapidly [4] , [5] . This autoregulatory effect implies the tight regulation of σG activation so that its normal timing and cell specificity are both observed , and raises questions regarding the mechanisms that prevent activation of the positive feedback in the forespore prior to engulfment completion , or in non-sporulating cells . Three negative regulators of σG are known , the LonA protease , and the anti-sigma factors SpoIIAB and CsfB [12] , [20]–[22] . LonA , an ATP-dependent serine protease , acts mainly to prevent inappropriate activity of σG under culture conditions in which sporulation is not favored [22]–[24] . During sporulation LonA may only be active in the mother cell , because its forced expression in the forespore strongly interferes with sporulation [23] , [24] . Genetic and biochemical experiments have shown that SpoIIAB , the anti-sigma factor that maintains σF inactive prior to the asymmetric division of sporulating cells , also binds to σG [12] , [24]–[26] . However , while SpoIIAB contributes to the inactivity of σG under non-sporulation conditions and in the mother cell during sporulation it does not play a critical role in the negative regulation of σG in the forespore [8] , [21] , [24] ) . A third negative regulator of σG is CsfB ( also known as Gin ) , a novel type of Zn2+ anti-sigma factor [20] , [27] , [28] . CsfB combines two properties expected for a factor capable of inhibiting σG prior to engulfment completion: specificity for σG ( unlike SpoIIAB , CsfB does not binds to σF ) and its early presence in the forespore compartment [20] , [28] , [29] . However , although one group initially proposed that CsfB had a key role in the negative regulation of σG in the pre-engulfed forespore [20] , other groups did not observe massive premature activation of σG in the forespore upon deletion of the csfB gene [8] , [27] . While the auto regulatory nature of σG seems to justify the existence of multiple negative regulators , none of the known regulators per se , seems to have a decisive role in preventing activation of the σG positive feedback loop . Because σF and σG are very similar proteins , we reasoned that the residues in which the two proteins differ could hold the key to their differential regulation . We changed all the residues within conserved regions 1 . 2 through the beginning of region 2 . 3 of σG in which it differs from σF to the residue found in this latter protein . We report the identification of a mutation ( N45E ) that reduces binding of CsfB to σG in vivo and in vitro . The mutation also results in the appearance of a population of stationary phase cells in which σG becomes active . We show that σG drives expression of csfB in these cells , setting-up a negative feedback loop that limits its activation across the population . We further show the importance of N45 in σG and its equivalent in σF ( E39 ) , in the different responsiveness of the two forespore-specific sigma factors to CsfB . While unable to bind to wild type σF , CsfB interacts with a form of σF in which E39 is replaced by an N residue , found in the corresponding position of σG ( N45 ) . Importantly , we show that the E39N substitution can strongly inhibit the forespore-specific activity of σF and the efficiency of sporulation . Thus , a single amino acid residue allows CsfB to discriminate between the two highly similar forespore-specific sigma factors . This property is likely to be widespread , because N45 is invariant in Bacillus orthologues of σG , while with a single exception N is excluded from the equivalent position in the σF proteins of the same species .
Since σF is active in the forespore in a temporal window when σG is kept inactive ( Figure 1A ) , we reasoned that we would be able to find one or more substitutions that would render σG prematurely active . We initiated this study by changing most of the residues within regions 1 . 2 and 2 . 1 through the beginning of region 2 . 3 of σG that differed from σF to the amino acid found at the equivalent position in this latter protein ( Figure 1B ) . The mutations were generated in vitro and transferred by congression to the sigG locus ( see the Materials and Methods section ) . We then screened for mutants exhibiting elevated levels of σG -directed gene expression under non-sporulation conditions ( during growth in LB ) as these conditions previously led to the identification of two negative regulators of σG [21] , [22] . This is possible because active σG utilizes its own promoter , leading to the establishment of a positive auto regulatory loop that reinforces its activity [5] . We found a single substitution at codon 45 of the sigG gene , an asparagine to a glutamic acid ( henceforth N45E ) that increased the activity of σG in vivo , as monitored using a fusion of the σG-responsive sspE promoter to lacZ [5] . The sspE gene codes for an abundant small acid-soluble protein required for the efficient return of spores to vegetative growth , and that is normally expressed in the forespore when σG becomes active [30]–[32] . The N45E mutation stimulated PsspE-lacZ transcription in colonies of cells growing on solid medium as well as in cells growing in liquid medium , where β-galactosidase activity was 2 fold higher in N45E mutant cells than in wild type cells ( Figure 2A and 2B ) . On liquid medium , the activity of σGN45E was higher when the cells entered stationary phase ( Figure 2B ) . The augmented expression of PsspE-lacZ could be due to increased activity of σG or alternatively to the titration by σGN45E of a negative regulator of σF , which at least under some conditions is also able to direct transcription from the sspE promoter [5] . To test the model that σGN45E could titrate an inhibitor of σF , we first examined the effect of two additional point mutations , F91A and Y94A , in region 2 . 3 of σG ( see Figure S1A ) . These residues are presumed to play a role in promoter melting ( reviewed by [33] ) , and alanine substitutions at these positions , while allowing the accumulation of σG , inactivate the sigma factor ( Text S1 and Figure S1 ) . Importantly , the N45E-stimulated expression of PsspE-lacZ was abolished in a N45E/F91A/Y94A triple mutant ( data not shown ) . This finding established that the N45E stimulated transcription of PsspE-lacZ was dependent on σG itself . None of the other sigG mutations screened increased expression of PsspE-lacZ , as illustrated by the sigGV44I mutant , bearing a valine to isoleucine substitution at codon 44 ( V44I ) ( Figure 2A and 2B ) . We hypothesized that the N45 residue was a contact site for a putative inhibitor of σG , which was eliminated by the N45E substitution . As a test of this idea we replaced the asparagine residue by an alanine ( henceforth N45A ) , a substitution expected to remove any positive contribution of the wild type amino acid side chain to a presumed interaction while maintaining protein structure [34] . Unexpectedly , the N45A substitution did not increase σG-directed transcription on colonies of cells growing on LB medium nor on liquid medium cultures ( Figure 2A and 2B ) . This observation suggests that the side chain of N45 may not be essential for a direct interaction of σG with an inhibitory factor . One alternative possibility is that N45E interferes with the binding of a putative inhibitor to σG . We next studied the effect of the N45E substitution on the activity of σG during sporulation in liquid Difco sporulation medium ( DSM ) . In this system , sporulation is induced by exhaustion of key nutrients , and its onset defined as the point at which a culture enters stationary phase . None of the sigG mutants that we screened in LB medium caused a Spo− phenotype ( data not shown ) , but we looked at PsspE-lacZ transcription during sporulation as the mutations could alter the normal activity profile of σG . In wild type cells , expression of PsspE-lacZ was sharply induced 4 hours after the onset of stationary phase and reached maximum levels around hour 6 ( Figure 2C ) . In keeping with the link between the activity of σG and engulfment completion , induction of PsspE-lacZ expression at hour 4 coincided with forespore engulfment in most cells of the population , as judged by FM4-64 staining ( not shown ) . In N45E cells PsspE-lacZ expression followed a bi-modal pattern , with an early period that peaked 2 hours after the onset of stationary phase and a second , starting at hour 4 , superimposable to the window of σG activity seen for wild type cells ( Figure 2C ) . The activity profile of σG and σGN45E paralleled the accumulation of the proteins , as assessed by immunobloting with an anti-σG antibody [17] . Both σG and σGN45E accumulated to maximum levels at hour 4 of sporulation in consonance with the main period of PsspE-lacZ expression , following engulfment completion ( Figure S2C ) . However , σGN45E begun to accumulate earlier than the wild type protein , soon after the onset of stationary phase in DSM , which correlates with the first period of PsspE-lacZ expression in the N45E mutant ( Figure S2C ) . Only the second period of PsspE-lacZ expression was seen for the V44I and the N45A mutants ( Figure 2C ) , consistent with the observation that these mutations did not enhance expression of the reporter fusion in our initial screen ( see above ) . Also consistent with the conclusion of our initial screen that the increased expression of the PsspE-lacZ reporter was not indirectly caused by titration of a negative regulator of σF ( see above ) , the N45E mutation did not increase expression of a lacZ fusion to the promoter for a gene , spoIIQ , controlled by σF ( spoIIQ-lacZ , [13] ) ( Text S1 and Figure S2A ) . In addition , the first period of σGN45E activity was still observed independently of sigF , coding for σF , which normally drives transcription of sigG in the forespore ( Figure 2F; see also below ) . While the sspE promoter can also be utilized by σF [8] , [27] , it is clear that the first period of σGN45E activity is σF-independent . This first period also occurred in cells with deletion mutations of the spoIIIJ ( Figure 2E ) and spoIIIA loci ( not shown ) , which are required for σG activity following engulffment completion . In fact , the first peak of σGN45E activity was seen even in cells of a spo0A deletion mutant ( [35] , [36] , and references therein ) , which codes for the master regulatory protein governing entry into sporulation and without which the asymmetric division that produces the forespore compartment does not takes place [37] ( Figure 2F ) . Altogether , these results show that the effect of the N45E substitution on PsspE-lacZ transcription during stationary phase in sporulation medium was dependent on and mediated by σG . The results also show that the second peak of σGN45E activity remained dependent on the normal control mechanisms that govern σG production and activation during sporulation . The results described in the preceding section could be explained if the N45E mutant segregated two distinct cellular populations , one with a normal pattern of σG activity , the other activating σG independently of sporulation . To test this possibility , the activity of σGN45E was localized during stationary phase in DSM , using a PsspE-cfp transcriptional fusion [10] . Note that under our experimental conditions , asymmetric division was completed in most of the cell population between hours 2 and 3 after entry into stationary phase ( as assayed by staining with the membrane dye FM4-64 ) , and engulfment was completed around hour 4 ( above ) . In agreement with previous results , expression of PsspE-cfp in wild type cells was only detected in the forespore at hour 4 after the onset of stationary phase , and only in cells in which the forespore had been engulfed by the mother cell ( Figure 3 and Table 1 ) . Note that no fluorescence was detected in cells of a sigG deletion mutant ( Figure 3 ) , confirming that the detected expression of the fluorescent reporter relied on σG . In the N45E mutant , however , around 1% of the cells scored between hour 0 and 2 after the onset of stationary phase showed strong whole-cell fluorescence ( Figure 3 and Table 1 ) . These cells had no morphological signs of sporulation , i . e . , they did not show asymmetric septa or engulfing membranes as assessed by FM4-64 staining . Consistent with the absence of asymmetric septation , we found that these cells did not show Pspo0A-yfp expression ( not shown ) and time-lapse microscopy experiments revealed that they eventually lysed ( Figure S3 ) . A second , larger population of N45E cells consisted of organisms that resembled the wild type in that they begun to display forespore-specific cfp fluorescence at hour 3 ( Figure 3; Table 1 ) . These cells did not show premature , whole-cell expression of PsspE-cfp . The results show that the first period of σG activity in the N45E mutant can be accounted for by a sub-population of cells that enter stationary phase and that do not enter in sporulation . We then focused our attention in the mechanism of activation of σGN45E in post-exponential phase cells . We considered the possibility that the N45E substitution made σG less responsive to the SpoIIAB anti-σG factor , which binds to and contributes to the negative regulation of σG in non-sporulating cells [8] , [12] , [24] . However , we found the activity of σGN45E to remain sensitive to SpoIIAB in vivo ( Text S1 and Figure S2B ) . While the possibility that the N45E substitution made σG refractory to SpoIIAB seemed discarded , the profile of σGN45E activity , in particular the first period of activity detected in stationary phase DSM cultures , was reminiscent of the effect reported for a mutation in csfB , which codes for the CsfB anti-σG factor [8] , [20] , [27] . For this reason , we examined the contribution of a csfB deletion mutation to the effect of the sigGN45E allele on σG-directed gene expression . On LB medium supplemented with X-Gal , the double mutant exhibited levels of β-galactosidase activity similar to the single sigGN45E or csfB mutants ( Figure 2A and 2B ) . In DSM the double mutant showed the bi-modal temporal pattern of PsspE-lacZ expression seen for the csfB or sigGN45E single mutants , but with β-galactosidase levels during the first period of expression higher than those of the sigGN45E mutant ( Figure 2D ) . There was no detectable effect of the mutations alone or in combination , on the second period of PsspE-lacZ activity ( Figure 2D ) . When examined by fluorescence microscopy , the sigGN45E/csfB double mutant resembled the single mutants: about 1% of the cells displayed early whole-cell fluorescence ( between hours 0 and 2 of sporulation ) whereas most of the population showed CFP fluorescence in the forespore following engulfment completion ( Figure 3 and Table 1 ) . Presumably , the fraction of pre-divisional cells with a strong whole-cell CFP signal corresponds to the β-galactosidase producing cells during the first hours of sporulation ( Figure 2C–2F ) . In conclusion , sigGN45E cells phenocopied the csfB mutant and the sigGN45E/csfB double mutant did not differ significantly from either single mutant . These findings suggest that the csfB and sigGN45E alleles exert their effect on σG by acting on the same pathway . The idea that both csfB and the sigGN45E allele act on the same pathway suggested to us that the N45E substitution could interfere with binding of CsfB to the mutant form of σG . In earlier work , CsfB and σG were found to directly interact in a yeast two-hybrid assay , and the first 71 residues of σG to be required for the CsfB-dependent inhibition of σG in vivo [20] . We used a similar approach to investigate whether σGN45E was less efficiently bound by CsfB . σG , σGN45A , σGN45E or CsfB were translationally fused to the C-terminus of the Gal4 DNA binding ( BD ) and activation domains ( AD ) , and the various fusion proteins expressed in different combinations in yeast cells and checked for their ability to interact in vivo , as assessed by the expression of a lacZ gene preceded by a Gal4-responsive element . As shown in Figure 4A and 4B ) , CsfB interacts efficiently with σG and only slightly less well with σGN45A . In contrast , CsfB interacts only weakly with σGN45E . We then used affinity chromatography to further investigate the interaction between CsfB and the different forms of σG . Whole cell extracts were prepared from cultures of a B . subtilis strain producing a functional CsfB-GFP fusion , 2 hours after the onset of sporulation , when σF is active and CsfB is known to accumulate [29] . The extracts were incubated with GST-σGwt , GST-σGN45A , GST-σGN45E or GST alone bound to glutathione agarose beads . Bound proteins were eluted and identified by immnunoblot with an anti-GFP antibody ( see Materials and Methods ) . These experiments showed that CsfB was pulled down efficiently by immobilized GST- σGwt but not by GST itself ( Figure 4C ) . GST-σGN45A pulled down CsfB-GFP less efficiently that the wild type ( the efficiency was 0 . 7× of the wild type ) but importantly , for σGN45E the efficiency of the pull down was about 0 . 4× of the wild type ( Figure 4C; note that the numbers in the panel represent averages for three independent experiments ) . We also note that in these assays SpoIIAB was pulled down by all forms of GST-σG with similar efficiency ( Figure 4C ) , suggesting that the N45A or N45E substitutions do not significantly affect binding of SpoIIAB to σG , and in line with the results of the in vivo activity experiments in which σGN45E was still susceptible to SpoIIAB ( see above; Figure S2A ) . To discard the possibility that the reduced retention of CsfB by σGN45E was caused by increased binding of a competing protein present in the B . subtilis extracts , the assay was repeated using a CsfB-Strep II-tagged protein overproduced and purified from E . coli . CsfB-Strep II was soluble when overproduced in a minimal medium only in the presence of Zn2+ , or in LB , which contains high levels of Zn2+ ( Text S1 and Figure S4 ) . The CsfB-Strep II protein purified from LB medium had bound Zn2+ ( metal to protein ratio of 1∶1 ) , as determined by atomic absorption spectroscopy . We incubated purified CsfB with GST or the various GST-σG forms immobilized on glutathione beads . After washing , CsfB was detected in the eluates by immunoblot with an anti-Strep II tag antibody . The CsfB protein was retained by GST-σG and by GST-σGN45A ( ∼0 . 6× the efficiency of the wild type ) , and to a lower level ( ∼0 . 2× of the wild type ) by GST-σGN45E ( Figure 4D ) . In these experiments , the signal in the pull down could be matched to that of a dilution of purified CsfB-Strep II ( Figure 4D ) . Although the differences between σG/σGN45A and σGN45E were more pronounced in the yeast two-hybrid experiments , both this assay and the pull-downs are in general agreement . Together , the results show that N45E is the substitution with the greatest impact on binding of CsfB to σG . csfB was first identified as a gene under the control of σF , and hence transcribed in the forespore soon after asymmetric septation [29] . Yet , in our hands , the main effect of a csfB deletion on the activity of σG activity was manifested in predivisional cells , i . e . , before the activation of σF ( Figure 3 ) . Previously , Chary et al . , ( 2007 ) have speculated that there is a basal level of σF-directed transcription during vegetative growth . However , our results suggest that the increased activity of σGN45E , which as we show is at least partially resistant to CsfB , was dependent solely on σG ( see above ) . Therefore , and although the expression of csfB in the forespore is not thought to be controlled by σG [38] , it seemed plausible that transcription of csfB in pre-divisional cells could be at least in part , controlled by σG . As a first test to this idea , we investigated whether expression of csfB and the activity of σGN45E co-localized . We first replaced the wild type csfB allele by a csfB-yfp fusion . This csfB-yfp fusion was subsequently transferred to strains carrying either the wild type or the N45E alleles of sigG and in addition , the σG reporter PsspE-cfp [10] . In the wt strain grown in DSM , organisms began to display forespore-specific yfp fluorescence between hour 1 and 2 after the onset of stationary phase ( Table 2 ) , consistent with the timing of septation and the activation of σF . No fluorescing organisms were observed before hour 1 , or in cells of a ΔsigF mutant , as expected for a σF-controlled gene ( Table 2; data not shown ) . Conversely , cfp fluorescence was only observed in the forespore 2 hours after the onset of sporulation , as expected for a σG-controlled gene and demonstrating the functionality of the csfB-yfp fusion . In the N45E mutant , between 1% to 4% of the bacteria displayed both whole-cell YFP and CFP fluorescence during the first hours of stationary phase in DSM ( Table 2 ) . Around hour 2 , the first cells showing forespore-specific expression of csfB-yfp were detected followed , around hour 3 , by cells with engulfed forespores showing PsspE-cfp expression . From this analysis , it is clear that the whole-cell expression of csfB-yfp early in stationary phase is confined to cells that also show activity of σGN45E , suggesting that csfB-yfp was transcribed under the direction of σG . As a further test to the possibility that σG controlled transcription of csfB , we made use of the sigG inactive allele described above in which the N45E mutation was combined with the F91A and Y94A “promoter-melting” mutations ( Figure S1 ) . In cells of the triple sigG mutant , no whole-cell YFP or CFP fluorescence was detected . In addition , and as expected , cells of the triple mutant did not display forespore-specific CFP fluorescence ( which is σG-dependent ) but showed forespore-specific YFP fluorescence ( which is σF-dependent ) ( Table 2 ) . These results strongly suggest that the expression of csfB in pre-divisional cells is σG-dependent , a conclusion reinforced by the observation that deletion of sigF in a N45E background abolished both the forespore-specific YFP and CFP fluorescence ( σF- and σG- dependent ) while maintaining the early whole-cell expression of yfp and cfp ( Table 2 ) . Lastly , as a more direct test for the ability of σG to control the expression of csfB , we monitored the expression of a PcsfB-lacZ fusion upon artificial induction of σG production from PxylA in vegetatively growing cells . The results in Figure 5A show that addition of xylose resulted in the induction of csfB-lacZ expression , even in the presence of a sigF deletion mutation , consistent with the view that σG can also drive expression of csfB , and with the similarity of the −10 and −35 promoter elements recognized by σF and σG [39]–[41] . Taken as a whole , the results suggest that the capacity of σG to drive production of CsfB in pre-divisional cells may be part of a mechanism to limit the ectopic activation of σG should any condition promote its activation . If production of CsfB is part of a regulatory circuit that self-restrains the activity of σG , then mutations in other factors known to negatively regulate σG should also induce expression of csfB , and the extent of the effect across the population should reflect the contribution of the affected regulator to the regulation of σG . Two such factors are known , the LonA protease and the SpoIIAB anti-sigma factor , which act independently to negatively regulate the activity of σG , mainly under non-sporulation conditions [21] , [22] . To determine the relative impact of mutations known to affect the regulation of σG on its activity across the population , and whether those mutations also increased the expression of csfB , we used fluorescence microscopy to simultaneously quantify the expression of PsspE-cfp and csfB-yfp at the onset of stationary phase in LB , in a panel of strains carrying PxylA fusions to wild type sigG , sigGN45E , sigGE156K ( coding for a form of σG refractory to SpoIIAB; [24] ) , sigGN45E/E156K or a PxylA-sigGwt construct in combination with a lonA deletion mutation ( Figure 5B ) . The growth medium was supplemented with 0 . 001% xylose , as in preliminary experiments ( Text S1 and Figure S5 ) this was the highest concentration at which wild type σG could be induced without causing significant cell lysis . In control experiments , no fluorescence could be detected in strains lacking either of the PsspE-cfp and csfB-yfp fusions ( not shown ) . The results in Figure 5C ( top graph ) show a clear correlation between the YFP and CFP signals for all strains tested . Cells that produce CFP also produce YFP , and an increase in the expression of one reporter is accompanied by an increase in the expression of the other ( Figure 5C , top ) , highlighting the link between the activity of σG and the production of its negative regulator , CsfB . The middle and lower graphs of Figure 5C are cumulative frequency distributions of the CFP and YFP signals for the various strains . For the N45E , lonA , and N45E/E156K strains , about 50% and 40% of the population shows CFP and YFP signals , respectively , above 2 arbitrary units . In contrast , only 10% of the wt or E156K populations show CFP or YFP signal intensities above this value ( Figure 5C ) . Induction of σGN45E/E156K increased the number of cells with high CFP fluorescence ( above 8 arbitrary units ) to 20% of the population , as compared to 10% for the strains bearing the single N45E , E156K or lonA mutations . This observation is in agreement with the idea ( see above ) that σGN45E is still sensitive to SpoIIAB . Smaller differences in the YFP signal distribution were seen between the double N45E/E156K mutant and the single N45E , E156K and lonA mutants , possibly reflecting reduced YFP stability . While CsfB , mainly , and LonA emerge as the principal regulators of σG activity during entry into the stationary phase of growth , SpoIIAB per se seems to have only a minor role ( Figure 5C ) . Importantly , we were unable to combine the sigGN45E/E156K allele with a lonA deletion , highlighting the convergent action of CsfB , SpoIIAB and LonA in the negative regulation of σG , and suggesting that these are likely to be the main , if not the only , negative regulators of σG at play . The results also unravel a negative σG autoregulatory loop ( Figure 5D ) , in which by commanding the expression of csfB , the fraction of cells with ectopic activity of σG is curtailed . Since production of σG in the strains above was driven from the PxylA promoter , we expected the various forms of σG to accumulate to similar levels , independently of the number of cells showing σG activity . This was verified by immunobloting analysis with an anti-σG antibody , for the N45E , E156K , N45E/E156K and wild type σG in the lonA background ( Figure 5E ) . Because changing N45 of σG for the residue found at the equivalent position in σF , E39 , makes σG less efficiently bound by CsfB , and since CsfB does not bind to σF [20] , we reasoned that perhaps this position was essential for the discrimination by CsfB between the two forespore-specific sigma factors in vivo . This inference was strengthened by the observation that N45 is invariant in σG orthologues , whereas E39 is highly conserved among σF proteins ( Figure 1C ) . Therefore , we decided to investigate whether E39 was important for the resistance of σF to CsfB , and for the regulation of its activity in vivo . We conducted GAL4-based yeast two-hybrid experiments to test the interaction between wild type σF and a mutant form of the protein with the E residue at position 39 replaced by an N ( E39N; Figure 1B ) . In agreement with the results of an earlier study [20] , CsfB did not interact with σF in our assay ( Figure 6A and 6B ) . In contrast , CsfB interacted efficiently with σFE39N ( Figure 6A and 6B ) . Thus , the E39N substitution is sufficient to allow binding of CsfB to σF . We next wanted to test whether the presence of an N at position 39 of σF , expected to make it susceptible to CsfB , would affect spore development . We found the E39N substitution to cause a 5-fold decrease in the efficiency of sporulation ( data not shown ) . Chary et al . found that when csfB is expressed from the IPTG-inducible Pspac ( Hy ) promoter prior to the activation of σF , spore formation was severely reduced [27] . However , in this strain the activity of σF was not impaired and spore formation was blocked in the developmental pathway just after engulfment completion [27] . We used a similar assay to test for the effect of the E39N mutation on the activity of σF . We transferred the IPTG-inducible Pspac ( Hy ) -csfB fusion to strains mutant for sigF and with a second copy of the entire sigF operon ( with either the wild type sigF cistron or sigFE39N ) integrated at the amyE locus under the control of its native promoter . In the strain carrying the wild type allele of sigF grown in the presence of IPTG to induce csfB expression prior to the activation of σF , spore formation showed the reported 103-fold reduction relative to cultures without IPTG ( Figure 6C ) confirming the results of Chary et al . ( 2007 ) . Strikingly , induction of csfB expression in the strain carrying the sigFE39N allele of sigF reduced spore formation 106-fold compared to the uninduced cultures ( Figure 6C ) . To investigate whether the more drastic sporulation defect observed in the strain carrying the E39N allele was due to impaired σF activity , we used a fusion of the σF-dependent yuiC promoter to gfp [42] . GFP fluorescence was monitored by microscopy during sporulation in the strains with IPTG inducible expression of csfB and bearing either the wild type or sigGE39N alleles . We found that the induction of csfB reduced the activity of σFE39N to 10% of the levels observed when CsfB was produced prior to asymmetric division in the presence of wild type σF ( Figure 6D ) . Thus , CsfB can interfere strongly with the activity of σFE39N in vivo .
The production of transcription factors often leads to the activation of gene expression during cell differentiation and development . In some instances , positive auto-regulation of the transcription factor drives gene expression in the differentiating cell down a specific developmental path . However , the power of these positive feedback loops raises the potential for inappropriate expression of the transcription factor in the wrong cell or at the wrong time . Therefore , the expression of autoregulatory transcriptional activators must be tightly controlled . We show here that CsfB has a function in preventing the activation of the forespore-specific , auto-regulatory σG factor , in stationary phase cells , in either a medium that does not support sporulation or in a sporulation medium , prior to the asymmetric division that initiates the program of compartment-specific gene expression that leads to differentiation of the spore . This role of CsfB was uncovered because the N45E substitution in σG , which reduces binding by CsfB , also results in activation of the sigma factor in a fraction of stationary phase cells . We show that CsfB is also produced , under σG control , in the same stationary phase cells where σGN45E becomes active . Hence , a negative feedback loop is established which , with the help of SpoIIAB and LonA , dominates the positive feedback loop involving σG , and keeps its activity low . The role of LonA and SpoIIAB in the negative regulation of σG in stationary phase cells was shown before [9] , [12] , [17] , [21] , [22] , but our analysis suggests that CsfB and LonA are the main regulators of σG . Nevertheless , the role of SpoIIAB is evidenced when the N45E and E156K substitutions are combined , and by our inability to construct a strain additionally carrying a lonA deletion . The lethality of this triple mutant further suggests that CsfB , SpoIIAB and LonA may be the only negative regulators of σG at play in stationary phase cells . Although mutations that interfere with the function of CsfB , SpoIIAB or LonA may cause strong expression of σG-dependent genes in pre-divisional cells , this only occurs in a fraction of the population ( Figure 5 ) . We do not presently know whether the cells which show σG activity are somehow different from the rest of the population at some fundamental level , or whether σG activity arises because of random fluctuations in the levels of σG itself , and its negative regulators . In any case , high-level expression of even wild type σG in stationary phase cells causes cell lysis , emphasizing the importance of limiting the potential for σG activation ( Figure S6 ) . Lysis may be a consequence of high levels of σG activity [26] , an indirect effect of the release of σF through titration of SpoIIAB by σG [20] or both , as induction of σG production in LB leads to lysis even in the absence of σF ( data not show ) . CsfB was initially proposed to be a key factor in keeping σG inactive in the forespore prior to engulfment completion [20] . However , a more consensual view of the role of CsfB is that the anti-sigma factor acts as a timing device , to help prevent σG activity prior to engulfment completion [8] , [9] , [27] , [38] . The results of our investigation are in line with this view , as in our hands deletion of csfB or the N45E substitution in σG ( which we show prevents binding of CsfB to σG ) did not bypass the genetic and morphological controls that link the activity of σG to engulfment completion . We postulate that during sporulation the negative feedback loops contributes to counter the positive feedback loop involving σG until CsfB is inactivated , or its synthesis is reduced by an unknown second regulator , or σG accumulation overwhelms that of CsfB . It is not known if CsfB is inactivated in the forespore following engulfment completion , but the anti-sigma factor seems to rapidly disappear from the forespore once σG becomes active ( our unpublished results ) . It is also likely that an additional factor prevents expression of csfB in the engulfed forespore . For example , σG drives production of SpoVT a forespore-specific transcription factor , which represses at least 27 σG-dependent transcriptional units [41] , [43] . SpoVT has a C-terminal GAF ( cGMP-specific and cGMP-stimulated phosphodiesterases , Anabaena adenylate cyclases , and Escherichia coli FhlA ) -like domain , which is essential to modulate the DNA-binding activity of the N-terminal domain , and may respond to nucleotides or other small molecules [44] . The accumulation of nucleotides in the engulfed forespore in turn , may be essential for the activity of σG and may depend on the action of the SpoIIIA-Q channel [9] . Two observations are consistent with the interpretation that the N45E substitution interferes with binding of CsfB to σG . First , wild type σG could pull down CsfB from extracts of B . subtilis in sporulation medium , but σGN45E did so less efficiently , a difference that was amplified when purified CsfB was used ( Figure 4; see also below ) . Second , CsfB interacted with wild type σG but not with σG N45E in a yeast two-hybrid system ( Figure 4 ) . CsfB was purified from E . coli cells as a C-terminal fusion to the Strep II tag because in vivo a CsfB-GFP fusion was fully functional ( this work ) . The CsfB-Strep II protein had Zn2+ bound with a stoichiometry of 1∶1 . The Zn2+ was released by oxidation of the protein with H2O2 , suggesting the involvement of the conserved Cys residues in CsfB in its coordination ( see Figure S4 ) . Recently , a MalE-CsfB fusion protein was purified from sporulating cells of B . subtilis with Zn2+ bound with a stoichiometry of 0 . 5 mol/mol [28] . Together with genetic data , this suggested that CsfB could act as a dimer ( or higher order multimer ) and possibly alternate between an active and an inactive state [28] . Importantly , the activity of σG was efficiently inhibited in E . coli cells , when co-produced with CsfB [28] , and the CsfB-Strep II protein purified from E . coli cells clearly discriminated σG and σGN45E in our pull-down assays ( Figure 4E ) . The N45 residue in σG may contribute to the interaction with CsfB . If so , however , this contact does not seem to be essential because the N45A substitution did not result in increased activity of σG in vivo , and caused only a small reduction in the ability of σG to interact with CsfB in yeast two-hybrid and pull-down assays ( Figure 4 ) . Additional mutagenesis studies may illuminate if and how the N45 residue contributes to the interaction with CsfB , and how the N45E substitution interferes with the interaction . CsfB is likely to contact σG at other positions , and these other contact sites are likely to be present in σF as well . First , because no other single mutation was found within the first 150 residues of σG that would affect its activity in vivo and second , because while incapable of binding to wild type σF ( [20]; this work ) , CsfB bound efficiently to σFE39N ( see also below ) . The location of the N45 residue within region 2 . 2 of σG and its role in permitting binding by CsfB , is also consistent with previous work in which a σF/G chimeric protein allowed the target for CsfB to be mapped within the first 77 residues of σG [20] . The N45 residue is invariant among σG orthologues of Bacillus species and related organisms but less conserved among the σG proteins of the Clostridia ( Figure 1C ) . These observations highlight the importance of N45 ( and homologous residues ) in a sub-group of sporeformers including B . subtilis and related organisms , in which σG is regulated by the anti-sigma factor CsfB . The observation that the N45E substitution reduces binding of CsfB to σG provides a plausible explanation for the increased activity of σGN45E in vivo . However , we cannot at present discard the possibility that the N45E alteration , which affects a residue positioned within conserved region 2 . 2 , also increases binding of σG to core RNA polymerase . The position homologous to N45 is often occupied by an acidic residue in proteins of the σ70 family of sigma factors ( the σG orthologues of Bacillus species and related organisms being a conspicuous exception ) , and in the crystal structure of the σ70-containing RNA polymerase holoenzyme from Thermus aquaticus [45] , E189 ( homologous to N45 in the σG protein of B . subtilis ) is involved , with other neighboring residues , in a direct contact with residue K159 in the β′subunit ( Text S1 and Figure S6 ) . An asparagine residue , as is found in σG , could also contribute to the interaction with β′ at this site . However , an acidic residue would most likely make a stronger , electrostatic , contribution to the interaction . This in turn suggests that the N45E substitution could also enhance the activity of σG by favoring its interaction with the β′subunit of core RNA polymerase . If so , then the regulation of σG activity in vivo could involve competition between CsfB and β′ for binding to σG . In any event , the possible contact involving N45 and β′ is in line with the view that one mechanism by which anti-sigma factors function is by occluding sigma-core binding interfaces [46] , [47] . Two of the mutations known to impair binding of SpoIIAB to σF map within region 2 . 2 , and mark residues that are conserved in σG ( [24] , [48]; see also Figure 1B ) . This suggests that SpoIIAB and CsfB may use partially overlapping interfaces in binding to σG and may explain the competition between the two anti-sigma factors for binding to σG under certain conditions [20] , [28] , [48] . However , σGN45E was still bound by SpoIIAB and was still susceptible to SpoIIAB in vivo . Therefore binding of SpoIIAB to σG does not seem to require the N45 residue . The strict conservancy of N45 among σG proteins of other Bacillus species and related organisms , together with its nearly absolute exclusion from orthologues of σF , suggests an important , conserved role for this residue . CsfB does not seem to negatively modulate the activity of σF , consistent with its inability to bind to this σ factor [20] , [27] , [28] , [38]; this work ) . Because the E39N substitution is sufficient to allow binding of CsfB to σF , the E39/N45 position in the σF/σG families of proteins seems critical for the discrimination by the CsfB anti-sigma factor . Perhaps strengthening this idea , the only exception to the rule that an N is excluded from the critical position in σF is B . clausii ( Figure 1C ) , but in this organism no csfB orthologue could be identified ( not shown ) . Recently , a protein related to CsfB , and termed Fin , was shown to inhibit the activity of σF and to play an important role in promoting the switch from σF to σG in the forespore [38] . It is possible that the N45/E39 residues help enforcing the specific regulation of σF by Fin and of σG by CsfB . It is not known whether σFE39N is susceptible to Fin . However , the E39N substitution did not seem to affect the activity of σF and caused only a 5 fold reduction in the efficiency of sporulation ( this work ) , whereas deletion of fin increased the window of expression of σF-dependent genes , and caused a 50-fold reduction in the efficiency of sporulation [38] . Perhaps then , σFE39N is still regulated by Fin . CsfB was also proposed recently , to antagonize low levels of σE resulting from inappropriate activation in the forespore , thus contributing to the confinement of its activity to the mother cell [49] . It is not yet known whether CsfB interacts with σE . However , if so , and because an acidic residue ( E ) is found at the position equivalent to N45 in σG , it follows that in the context of the σE protein binding by CsfB is likely to involve other residues .
The B . subtilis strains used in this work are congenic derivatives of the Spo+ strain MB24 ( trpC2 metC3 ) , and are listed in Table S1 . The plasmids used in strain construction are described in the sections below and in Text S1 . LB medium was used for growth or maintenance of E . coli and B . subtilis , and sporulation was induced by growth and exhaustion in Difco sporulation medium ( DSM ) [24] . The Quick Change site-directed mutagenesis system ( Stratagene ) was used for the generation of all site-specific mutations , which were always confirmed by sequencing . We used pMS45 , containing the sigG gene [24] and sigG-specific primers ( all primers are listed in Table S2 ) to convert the residues highlighted in Figure 1B ( orange and red circles ) into the aminoacid found in σF . The various mutations were then transferred to the sigG locus by congression as it has been observed that certain mutations cause a greater increase in the activity of σG when sigG is inserted at an heterologous locus such as amyE ( [27]; our unpublished observations ) . For congression , pMS45 and its derivatives carrying the different sigG mutations , together with chromosomal DNA from AH6566 ( ΔsigG ΔyycR:: PsspE-cfp ΔsspE::PsspE-lacZ ) , was used to co-transform strain AH2452 ( ΔsigG ΔsspE::PsspE-lacZ ) with selection to CmR . Spo+ congressants appeared at a frequency of about 3% . One congressant for each sigG allele for which the presence of the desired mutation was confirmed by PCR and sequencing was kept for further study . Mutants that showed increased b-galactosidase production from the PsspE-lacZ reporter fusion were identified on LB plates containing 5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside ( X-Gal ) . The construction of fusions of the xylose-inducible xylA promoter to different sigG alleles , and of csfB-gfp , -yfp and lacZ fusions is described in detail in Text S1 , accompanying this article . First , the entire spoIIA operon was PCR amplified with primers sigF219D and sigF2032R ( Table S2 lists all primers used in this study ) , the 1813 bp product digested with BamHI and HindIII and inserted between the same sites of pDG364 [50] , This created pMS393 . Next , primers sigFE39ND and sigFE39NR were used to substitute the glutamate codon at position 39 of the sigF gene in pMS393 by an asparagine codon , which resulted in pMS394 . The coding regions of sigG , sigF and csfB were PCR amplified with primers sigG2016D and sigG2862R , sigF493D and sigF1318R , and primers csfB191D and csfB480R . The sigG and sigF PCR products were digested with NcoI and EcoRI and inserted between the same sites of pAS2-1 ( Clontech ) yielding plasmids pMS358 and pMS357 , respectively . We used pMS358 and primers sigGN45ED and sigGN45ER to substitute the asparagine codon at position 45 of σG by a glutamate codon . This resulted in plasmid pMS360 . We used pMS358 and primers sigGN45AD and sigGN45AR to substitute the asparagine codon at position 45 of σG by an alanine codon . This resulted in plasmid pMS429 . We used pMS357 and primers sigFE39ND and sigFE39NR to substitute the glutamate codon at position 39 of σF by an asparagine codon . This resulted in plasmid pMS387 . The csfB PCR product was digested with NcoI and SalI and inserted between the same sites of pACT2 ( Clontech ) yielding plasmid pMS356 . Mating of Sacharomyces cerevisiae strains and detection of β-galactosidase activity were performed as described before [51] . The spoIIAB coding region was PCR amplified with primers spoIIAB189D and spoIIAB698R . The PCR product was digested with BamHI and XhoI and inserted between the same sites of pET30a ( + ) ( Novagen ) creating pMS111 , in which the sequence for the His6 tag was introduced between the first and second codons of spoIIAB . pMS111 was introduced into competent cells of BL21 ( DE3 ) pLysS ( Novagen ) . Growth , induction , and lysate preparation was essentially as described [52] . The His6-SpoIIAB fusion protein was partially purified on His-Trap chelating columns as described by the manufacturer ( Amersham Pharmacia Biotech ) and used to raise a polyclonal anti-SpoIIAB antibody in rabbits ( Eurogentec , Belgium ) . First , primers csfB191D and csfBstrepR , which include the sequence coding for the Strep II tag ( IBA GmbH ) were used to PCR amplify the coding region of csfB . The resulting PCR product was digested with NcoI and BamHI and cloned between the same sites of pET16b ( Novagen ) to create pMS350 , which was then transformed into E . coli strain BL21 ( DE3 ) . The E . coli expression strain was grown to mid-log phase in LB ( 0 . 6 optical density at 600 nm ) , induced with 1 mM isopropyl-D-thiogalactopyranoside ( IPTG ) , and grown for 3 h before harvesting the cells . The cell pellets were resuspended in 3 ml portions of buffer A ( 100 mM NaCl , 10 mM Tris pH 8 . 0 , 10% glycerol ) per 50 ml of induced culture and lysed in a French pressure cell ( 18 , 000 lb/in2 ) . The lysate was centrifuged to remove cell debris . CsfB-Strep II tag was purified on Strep-Tactin Sepharose columns following the manufacturer instructions ( IBA GmbH ) . The metal content of the purified protein was analyzed by atomic absorption . Primers sigG2016D and sigG2964R were used to PCR amplify the coding regions of sigG , sigGN45A and sigGN45E from pMS45 , pJS4 , and pJS2 ( see above ) . The PCR products were digested with BglII and XhoI and cloned between the BamHI and XhoI sites of pGex4T-3 ( GE Healthcare ) to create pMS375 , pMS376 , and pMS377 , respectively , which bear in-frame N-terminal GST fusions to the different forms of σG . Derivatives of BL21 ( DE3 ) bearing each of these plasmids or pGex4T-3 ( GST-alone ) were grown to mid-log phase ( O . D . 600≈0 . 6 ) in LB , and induced with 1 mM IPTG for 3 h before the cells were harvested . The cell pellets were resuspended in 1 ml portions of buffer A [100 mM NaCl , 10 mM Tris-HCl ( pH 8 . 0 ) , 10% glycerol] per 50 ml of induced culture and lysed in a French pressure cell ( 18 , 000 lb/in2 ) . The lysate was cleared by centrifugation . One milliliter of cleared lysate was bound to 50 µl of a 50% slurry of glutathione Sepharose beads ( GE Healthcare ) at room temperature for 30 min . The beads were washed three times in buffer B ( same as A but with 200 mM NaCl ) . For the SpoIIAB and CsfB interaction assays , 1 ml portions of soluble extracts prepared from cultures of B . subtilis AH6608 ( csfB::km ΔsigG ΔamyE::csfB-gfp ) 2 h after the onset of sporulation were incubated for 30 min at room temperature with GST or the various GST fusions proteins bound to glutathione Sepharose beads or with the beads alone . The mixtures were washed three times with buffer B ( above ) , resuspended in a final volume of 30 µl , and subjected to SDS-PAGE and immunoblotting . Rabbit anti-GFP ( A . L . Isidro and A . O . Henriques , unpublished ) and anti-SpoIIAB ( above ) antibodies were used at dilutions of 1∶1000 and 1∶500 , respectively . An anti-σG antibody , at a 1∶1000 dilution , was used to control for the level of the GST-σG fusions immobilized [24] . For the CsfB interaction assay , 100 nM of purified CsfB-Strep II tag was incubated for 30 min at room temperature with the glutathione Sepharose beads complexed with the GST fusion proteins or with glutathione Sepharose beads alone . The mixtures were washed three times with buffer B ( above ) and resuspended in a final volume of 30 µl . The samples were subjected to SDS-PAGE and immunoblotting . An anti-Strep II tag polyclonal antibody was used at a 1∶1000 dilution ( IBA GmbH ) . For graphical representation of the data , the immunoblots were scanned and analyzed using the ImageJ software ( http://rsbweb . nih . gov/ij ) . Immunoblot analysis was used to monitor the accumulation of σG during growth or sporulation as previously described [17] . An anti-σA antibody was used as described before [36] . β-Galactosidadse activity was assayed with the substrate o-nitro-β-D-galactopyranoside ( ONPG ) , with enzyme activity expressed in Miller units [52] . Samples ( 0 . 6 ml ) of LB or DSM cultures were collected , resuspended in 0 . 2 ml of phosphate-buffered saline ( PBS ) and the membrane dye FM4-64 ( Molecular Probes ) added to a final concentration of 10 µg ml−1 . Microscopy was carried out as described previously [53] . Quantitative analysis of fluorescence intensity was done using the MetaMorph software package ( MDS Analytical Technologies ) . Data was analyzed and plotted using the “R” statistical computing and graphics software package ( www . r-project . org ) . | Positive auto-regulation of a transcriptional activator during cell differentiation or development often allows the rapid and robust deployment of cell- and stage-specific genes and the routing of the differentiating cell down a specific path . Positive auto-regulation however , raises the potential for inappropriate activity of the transcription factor . Here we unravel the role of a previously characterized anti-sigma factor , CsfB , in a negative feedback loop that prevents ectopic expression of the sporulation-specific sigma factor σG of Bacillus subtilis . σG is activated in the forespore , one of the two chambers of the developing cell , at an intermediate stage in spore development . Once active , a positive feedback loop allows the rapid accumulation of σG . Synthesis of both σG and CsfB is under the control of the early forespore regulator σF , and CsfB may help prevent the premature activity of σG in the forespore . However , CsfB is also produced under σG control in non-sporulating cells , setting a negative feedback loop that we show limits its ectopic activation . We further show that an asparagine residue conserved among σG orthologues is critical for binding and inhibition by CsfB , whereas the exclusion of asparagine from the homologous position in σF confers immunity to CsfB . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"cell",
"fate",
"determination",
"genetic",
"screens",
"gene",
"expression",
"genetics",
"biology",
"microbial",
"growth",
"and",
"development",
"microbiology",
"genetics",
"and",
"genomics",
"cell",
"differentiation",
"dna",
"transcription",
... | 2011 | A Negative Feedback Loop That Limits the Ectopic Activation of a Cell Type–Specific Sporulation Sigma Factor of Bacillus subtilis |
Restrictions on nematicide usage underscore the need for novel control strategies for plant pathogenic nematodes such as Globodera pallida ( potato cyst nematode ) that impose a significant economic burden on plant cultivation activities . The nematode neuropeptide signalling system is an attractive resource for novel control targets as it plays a critical role in sensory and motor functions . The FMRFamide-like peptides ( FLPs ) form the largest and most diverse family of neuropeptides in invertebrates , and are structurally conserved across nematode species , highlighting the utility of the FLPergic system as a broad-spectrum control target . flp-32 is expressed widely across nematode species . This study investigates the role of flp-32 in G . pallida and shows that: ( i ) Gp-flp-32 encodes the peptide AMRNALVRFamide; ( ii ) Gp-flp-32 is expressed in the brain and ventral nerve cord of G . pallida; ( iii ) migration rate increases in Gp-flp-32-silenced worms; ( iv ) the ability of G . pallida to infect potato plant root systems is enhanced in Gp-flp-32-silenced worms; ( v ) a novel putative Gp-flp-32 receptor ( Gp-flp-32R ) is expressed in G . pallida; and , ( vi ) Gp-flp-32R-silenced worms also display an increase in migration rate . This work demonstrates that Gp-flp-32 plays an intrinsic role in the modulation of locomotory behaviour in G . pallida and putatively interacts with at least one novel G-protein coupled receptor ( Gp-flp-32R ) . This is the first functional characterisation of a parasitic nematode FLP-GPCR .
Plant pathogenic nematodes ( PPNs ) impose a significant economic burden on global crop cultivation resulting in estimated losses of at least $118 billion per year [1] . The control of PPNs relies heavily on nematicides , basic crop rotation approaches , and the use of resistant crop cultivars; significantly many nematicides have diminishing utility as a consequence of their environmental toxicity . Consequently , global crop production remains under threat from PPNs for which no effective management strategies currently exist . While the PPN problem results from an absence of effective , legal control methods , deficiencies in animal and human parasite therapies are associated with escalating reports of resistance to chemotherapeutics [2] . Thus , while focused control of PPNs alone has significant merit , the identification of broad spectrum drug targets and chemotherapies which combat diverse nematode infections is highly desirable . Many of the most useful anti-nematode chemotherapeutics target neuromuscular signalling , compromising nerve-muscle function to impair normal parasite biology . One approach to novel drug target discovery in nematodes is the interrogation of alternative , unexploited , facets of this already proven repository ( see [3] for review ) . Within the neuropeptidergic system , motor functions ( reproduction , feeding and locomotion ) are known to be modulated by FMRFamide-like peptides ( FLPs ) ; the nematode FLPergic system remains unexploited for parasite control ( see [3]–[9] for review ) . The potential of FLPergic signalling as a control target resource embedded within nematode neuromuscular functionality has been highlighted [3] . FLP structural diversity hinges upon a conserved C-terminal tetrapeptide motif X-Xo –Arg-Phe-NH2 ( X represents any amino acid and Xo any hydrophobic amino acid except cysteine; [10] ) . BLAST-based bioinformatics has revealed a high degree of inter-species FLP conservation across nematode clades , with a number of species possessing a FLP complement with comparable complexity to the model nematode C . elegans , which expresses 31 flps encoding >70 distinct peptides [10]–[12] . If FLP structural conservation is mirrored by conserved function then the potential broad-spectrum utility of drugs directed against FLP signalling targets would be enhanced . While FLPs themselves are of limited control value , they do facilitate the identification of appealing targets , such as the G-protein coupled receptors ( GPCRs ) by which the majority mediate their biological effects . Although 12 FLP GPCRs have been deorphanised in C . elegans [13] , [14] , none have been characterised in nematode parasites; whilst flp-1 , -8 , and -18 peptides have been shown to interact with a latrophillin-like GPCR in Haemonchus contortus , their affinity for this receptor was low [15] . The C . elegans VRFamide receptor 1 ( C26F1 . 6 ) is potently activated by two peptides , TPMQRSSMVRFamide and AMRNALVRFamide [13] . In C . elegans a single copy of AMRNALVRFamide is encoded by flp-11 and flp-32 , however in G . pallida this peptide is encoded on flp-32 only [10] . Subsequent BLAST interrogation of available nematode genome , transcriptome and expressed sequence tag ( EST ) datasets has revealed that flp-32 is conserved in at least 16 nematode species , across two nematode clades , encompassing a number of contrasting lifestyles ( [11]; unpublished data ) . Such conservation suggests that FLP-32 may modulate functionally important signalling pathways within the neuromuscular signalling system; building a pan-phylum picture of flp-32 biology in multiple pathogenic nematode species will generate valuable information on the biological role of this peptide and validate its potential as a broad spectrum control target . Here we report the functional characterisation of Gp-flp-32 and a putative Gp-flp-32 receptor ( Gp-flp-32R ) from the potato cyst nematode ( PCN ) G . pallida , a pathogenic nematode which is readily amenable to reverse genetic techniques [7] , and boasts a completed genome sequence ( http://www . sanger . ac . uk/cgi-bin/blast/submitblast/g_pallida ) . We also describe a novel in vivo reverse genetics approach to putative FLP receptor deorphanisation in parasitic nematodes .
flp-32 is expressed in at least 16 nematode species where it encodes a highly conserved peptide with a characteristic VRFamide C-terminal motif , AMRN ( A/S ) LVRFG ( see Fig . 1B ) . Previous interrogation of G . pallida ESTs [11] identified a transcript encoding a putative FLP-32-like peptide ( GenBank accession number CV578361 ) , which was used in this study to aid PCR confirmation of the full length Gp-flp-32 transcript . Primers designed to confirm the open reading frame of Gp-flp-32 generated a 321 nucleotide cDNA sequence ( GenBank accession number JQ685131 ) , encoding a 107 amino acid ( aa ) protein ( Fig . 1A ) . The confirmed Gp-flp-32 aa sequence encodes a single copy of the FLP-32 peptide , AMRNALVRFG , flanked at both ends by dibasic residues ( KK/KR ) , and a 28 aa signal peptide ( see Fig . 1A; [16] ) . Further interrogation of the G . pallida EST database ( GenBank ) and genome assembly ( Wellcome Trust Sanger Institute , G . pallida November 2010 supercontig assembly ) in April-August 2011 did not reveal additional AMRNALVRFG encoding transcripts . Gp-flp-32 expression , visualised by the hybridisation of a 201 base pair ( bp ) probe , was identified both within and connecting the anterior and posterior regions of the nematode ( see Fig . 2A–D ) . Staining was evident in the circumpharyngeal nerve ring ( CNR ) , and within multiple distinct cell bodies in the ventral nerve cord ( VNC ) and lumbar ganglia ( LG ) ( see Fig . 2A–D ) . Staining within the CNR was diffuse with no specific neuronal cell bodies staining strongly ( Fig . 2A ) ; this pattern was evident in the majority of specimens ( >90% ) treated with the antisense probe , and is similar to a diffuse ISH staining pattern previously reported in the G . pallida CNR for flp-6 ( KSAYMRFG; [17] ) . In contrast , staining in the VNC was characterised by groups of three to four distinct and strongly reactive cell bodies spaced at regular intervals along the nerve cord , beginning posterior to the CNR and running into the tail ( Fig . 2B ) . Although slightly variable , most specimens exhibited approximately six to eight groups of cells in the VNC , with as many as 18 cell bodies visible at any one time ( see Fig . 2B and C ) . While unequivocal assignment of neuronal cell bodies is difficult , in this scenario it is likely that they belong to VD and/or DD motor neurons which possess cell bodies in the VNC of C . elegans , where they have been shown to express a VRFamide-like flp gene encoding peptides similar to that encoded by Gp-flp-32 [18] . VD neurons are a set of 13 motor neurons which innervate ventral muscle and have cell bodies in the VNC , while DD are a set of six motor neurons ( pre-synaptic to VD ) , which also possess cell bodies in the VNC , but instead innervate dorsal muscle [19] , [20] . Together this amounts to 19 identifiable cell bodies within the VNC , and while this number is known to vary in C . elegans according to developmental stage [19] , it is similar to the 18 cell bodies visible within the VNC of G . pallida following Gp-flp-32 antisense probe hybridisation . Defined staining was also identified in a tightly associated group of three cell bodies close to the tip of the tail in the region of the LG , a group of cell bodies which cluster together posterior to the pre-anal ganglia ( PAG ) and the termination of the VNC ( see Fig . 2D ) . Again whilst unequivocal identification of cells is difficult in such a tightly packed ganglion , C . elegans neurons which possess cell bodies in this region include: two cell bodies of the PVC interneuron ( PVCL and PVCR ) which are post-synaptic to VD motor neurons in the VNC and are known to regulate locomotion; and the motor/interneuron DVB which located in the dorso-rectal ganglion and sends commissures into the PAG and VNC . No staining was observed in negative control experiments . A custom raised antiserum directed against the single peptide encoded by Gp-flp-32 , AMRNALVRFamide , was used to localise Gp-FLP-32 using ICC in G . pallida J2s . The overall pattern of Gp-FLP-32 localisation was similar to the expression pattern of Gp-flp-32 exhibited in ISH experiments , comprising extensive AMRNALVRFamide immunostaining within the nervous system of G . pallida ( see Fig . 3A and B ) . Strong immunoreactivity was visualised within the CNR , with AMRNALVRFamide-immunopositive nerve processes running both anteriorly and posteriorly from the CNR towards the stylet protractor muscles and the VNC respectively ( Fig . 3A ) . While some staining within the CNR was diffuse as previously noted in ISH experiments , there were distinct accumulations of immunoreactivity on both the ventral and dorsal sides of the nerve ring ( see Fig . 3A ) . In addition there was an accumulation of immunopositive staining , albeit weaker than the anterior staining , in the posterior of the nematode close to the tip of the tail ( see Fig . 3B ) . Again this was concurrent with the positioning of Gp-flp-32 expression in the ISH experiments , which would suggest Gp-FLP-32 immunoreactivity in cell bodies of the LG . In this study , Gp-flp-32 expression was demonstrated in key neuronal processes involved in nematode motor-control , and when compared to technique-matched ISH data previously published for Gp-flp-6 , -12 , -14 and -18 [17] , Gp-flp-32 expression is much more extensive . As such , it is not unreasonable to suggest that Gp-flp-32 plays a broad role in the neuronal control of G . pallida motor function . Here we used RNAi soaking experiments , measurements of post-silencing changes in Gp-flp-32 transcript levels , and bioassays to assess nematode phenotype , in an attempt to elucidate the role of flp-32 in PCN . Consistent and statistically significant reduction in target transcript ( quantified as ΔΔCt of Gp-flp-32 transcript relative to Gp-ace reference transcript ) of 55 . 1±4 . 6% ( n = 3 ) was achieved in Gp-flp-32 siRNA treated worms when compared to untreated worms ( P<0 . 001 , q = 8 . 988 ) and non-native control siRNA treated worms ( P<0 . 001 , q = 9 . 716; see Fig . 4A ) . Post-RNAi , worm phenotype was assessed by visual observation; worms in all control treatments appeared normal . However , Gp-flp-32 siRNA treated worms exhibited an increased frequency of normal sinusoidal movement whereby they appeared to move faster than control worms . This phenotype was quantified through employment of a sand column migration time-course assay [7] , where worms were counted every 2 hours ( h ) as they migrated down a vertical sand column during a 6–8 h period . This demonstrated that Gp-flp-32 siRNA treated worms migrated significantly faster than untreated worms ( 2 h , 53 . 2±9 . 7% vs 20 . 1±2 . 9% migration respectively , P<0 . 001; 4 h , 84 . 3±6 . 3% vs 58 . 7±5 . 9% migration respectively , P<0 . 01; n = 3; see Fig . 4B and C ) and control siRNA treated worms ( 2 h , 53 . 2±9 . 7% vs 18 . 8±2 . 4% migration respectively , P<0 . 001; 4 h , 84 . 3±6 . 3% vs 63 . 6±3 . 6% migration respectively , P<0 . 01; n = 3; see Fig . 4B and C ) . At the 6 h migration time point fewer untreated ( 83 . 7±3 . 2% ) , and non-native control siRNA treated ( 86 . 6±2 . 0% ) worms had successfully migrated relative to Gp-flp-32 siRNA treated worms; control worms took a further 2 h to complete migration . . During the migration experiment , untreated and control siRNA treated worm migration did not differ significantly at any time ( P>0 . 05; n = 6; see Fig . 4C ) . Together these data suggest that Gp-flp-32 expresses an inhibitory neuropeptide , which , when silenced , induces an increase in locomotory activity . The marked stimulation of J2 migration rate following Gp-flp-32 silencing was achieved with only a 55% reduction in transcript , suggesting that the encoded FLP has profound depressive effects on locomotion in wild type worms . These data differ from all published RNAi studies on flp gene function in PPNs which are characterized by the induction of phenotypes encompassing unusual body posture , slower movement and/or paralysis [7] , [21] . To ascertain if this increased rate of migration in response to flp-32 silencing is mirrored by other PPNs , the same migration experiments were performed on the pre-parasitic J2 stage of the root knot nematode Meloidogyne incognita . In these experiments worms were pre-treated with an siRNA targeting M . incognita flp-32 ( Mi-flp-32; GenBank accession number CN443314; [11] ) with controls as described above . flp-32 silenced pre-parasitic M . incognita migrate more rapidly than untreated or siRNA control treated worms , mirroring the phenotype of Gp-flp-32 silencing G . pallida J2s ( see Fig . 5 ) . These data show that flp-32 plays a key role in the modulation of normal locomotory behaviour in pre-parasitic J2s of two major groups of plant endoparasitic nematodes . These observations are consistent with the hypothesis that FLP-32 depresses locomotory behaviour in wild-type pre-parasitic J2s . The role of Gp-flp-32 is consistent with its expression pattern; Gp-flp-32 was identified along the VNC in the cell bodies of DD and VD-like motor neurons , which in C . elegans control sinusoidal movement through their innervation of dorsal and ventral muscles , respectively [19] , [20] . DD motor neurons relax dorsal muscles during ventral muscle contraction [22] , and are believed to regulate the wave amplitude of sinusoidal movement [23] . The DD and VD motor neurons , responsible for the relaxation of dorsal and ventral muscles during sinusoidal movement , could do so in part due to the action of FLP-32 . While the evidence presented here strongly suggests a locomotory role for Gp-flp-32 , this does not discount the possibility that Gp-flp-32 may also regulate other processes in PPNs . Many of the currently available in vitro assays for post-RNAi phenotype analysis in PPNs are designed to assess the ability of worms to migrate and move normally , such that disruption to processes such as egg laying or larval development would not be recorded . With this in mind , we employed a potato plant infection assay to determine if Gp-flp-32 silenced worms displayed infection-associated phenotypes . To probe the function of Gp-flp-32 further , the infectivity of Gp-flp-32 silenced J2s was compared in a small-scale potato plant infection assay . A positive control siRNA [directed against transcript encoding acetylcholinesterase ( Gp-ace ) ; GenBank accession number FJ499505] which displays reduced locomotory activity post-RNAi ( unpublished data ) was employed , in addition to the standard untreated and non-native siRNA controls described above . The purpose of this positive siRNA control was to demonstrate the effect of reduced locomotory ability on nematode infection rate . RNAi treated nematodes were applied to the sand covering the root network of 2 week old potato plants , and after a period of 4 days , plant roots were analysed for the presence of nematodes . Gp-flp-32 siRNA treated worms displayed a significantly higher mean infection rate of 74 . 4±5 . 0% ( n = 4 ) compared to untreated ( 35 . 4±3 . 4% , P<0 . 001 , q = 7 . 72; n = 8 ) , non-native control siRNA ( 30 . 8±7 . 1% , P<0 . 001 , q = 8 . 18; n = 6 ) , and positive control siRNA ( 15 . 4±2 . 5% , P<0 . 001 , q = 10 . 67; n = 5 ) treated worms ( see Fig . 6 ) . When compared , the infection rates of untreated and non-native siRNA treated controls were not significantly different ( 35 . 4±3 . 4% vs 30 . 8±7 . 1% infection , respectively; P>0 . 05 , q = 1 . 03; Fig . 6 ) . This assay confirmed that the increased migration rate displayed by Gp-flp-32 silenced worms translated to increased plant root infection rate , i . e . the worms migrate to the root faster and/or infect the root more quickly . Whilst this may reflect an enhancement in sensory ability further improving the chances of host location success , this is unlikely since Gp-flp-32 is not localised in areas of the worm associated with chemoreception . The nature of the Gp-flp-32 RNAi phenotype raises a question regarding the ability of worms to sustain their migratory and invasion activities . It is possible that increased rates of migration and infection would more rapidly deplete the finite energy reserves in these non-feeding J2s , resulting in premature death . This possibility was investigated using Oil Red O lipid staining [24] , [25] and assessment of lifespan in Gp-flp-32 silenced J2s over a 14 day period following RNAi . This assay did not reveal increase in lipid depletion or death rates in Gp-flp-32-RNAi treated worms compared to controls ( data not shown ) . Database mining facilitated the identification of a putative FLP-32 receptor in G . pallida , orthologous to the C . elegans VRFa receptor R1 ( C26F1 . 6; see Fig . 7A ) . RACE PCR and sequencing confirmed the sequence of the putative G . pallida flp-32 receptor ( Gp-flp-32R ) . Primers designed to confirm the open reading frame of Gp-flp-32R generated a 1 , 170 nucleotide cDNA sequence ( GenBank accession number JQ685132 ) , encoding a 389 aa protein ( Fig . 7A ) . Gp-flp-32R encodes seven transmembrane helices and conserved residues at positions 48 , 52 , 76 , 80 , 136–138 , 222 , 273 and 318–320; the 136–138 sequence ( DRF ) is a common variation on the DRY motif at the cytosolic end of the third transmembrane helix and is common to rhodopsin-like GPCRs ( see Fig . 7A ) . In a reciprocal tBLASTn search of the C . elegans non-redundant nucleotide and protein database , C . elegans C26F1 . 6 was returned as the top scoring hit ( 53% identity to Gp-flp-32R ) . Further interrogation of the G . pallida EST database ( GenBank ) and genome assembly ( Wellcome Trust Sanger Institute November 2010 supercontig assembly ) between August and October 2011 with C . elegans C26F1 . 6 did not reveal additional homologous transcripts . BLAST searches of all available nematode EST , genomic and transcriptomic resources identified 12 C26F1 . 6 GPCR homologues from the 16 species which express FLP-32 encoding transcripts , spanning clades IV and V ( Fig . 7B ) . Until now orthologues of the deorphanised C . elegans FLP GPCRs have not been reported in a parasitic nematode . As a result of the lack of information regarding parasitic nematode neuropeptide receptors , C . elegans represents the sole and limited source of GPCR functional data available for nematodes . Nevertheless , combining C . elegans receptor-ligand pairing data with sequence data homologies , appropriate expression patterns and matching RNAi phenotypic readouts can facilitate the functional characterisation of peptide-GPCR relationships in parasites . The identification of a putative Gp-flp-32R candidate has facilitated the application of functional characterisation tools to ( i ) confirm the identity of Gp-flp-32R as a FLP-32 activated receptor , and ( ii ) further probe the neuromodulatory role of Gp-flp-32 and Gp-flp-32R in G . pallida . To achieve this we duplicated the RNAi , qPCR , and migration bioassay experiments previously employed for Gp-flp-32 characterisation on Gp-flp-32R . Consistent and statistically significant reduction in target transcript ( quantified as ΔΔCt of Gp-flp-32R transcript relative to Gp-ace reference transcript ) of 75 . 7±5 . 6% ( n = 3 ) was achieved in Gp-flp-32R siRNA treated worms when compared to untreated worms ( P<0 . 001 , q = 9 . 459 ) and non-native control siRNA treated worms ( P<0 . 001 , q = 10 . 02; see Fig . 4A ) . Following RNAi experiments all worms appeared normal except that Gp-flp-32R silenced worms appeared to show an increased frequency of normal sinusoidal movement compared to controls . This phenotype matched that displayed by Gp-flp-32 silenced worms . To further probe the nature of this phenotype , sand column migration assays were employed to monitor worm migration every 2 h over a 6-h period . Phenotype analysis by migration assay showed that Gp-flp-32R siRNA treated worms migrated significantly faster than untreated worms ( 2 h , 57 . 1±13 . 9% vs 20 . 1±2 . 9% migration respectively , P<0 . 001; 4 h , 90 . 9±4 . 4% vs 58 . 7±5 . 9% migration respectively , P<0 . 001; n = 3; see Fig . 4B and C ) and control siRNA treated worms ( 2 h , 57 . 1±13 . 9% vs 18 . 7±2 . 4% migration respectively , P<0 . 001; 4 h , 90 . 9±4 . 4% vs 63 . 6±3 . 6% migration respectively , P<0 . 001; n = 3; see Fig . 4B and C ) for the first 4 h of migration . At the 6 h migration time point fewer untreated ( 83 . 7±3 . 2% ) , and non-native control siRNA treated ( 86 . 6±2 . 0% ) worms had successfully migrated relative to Gp-flp-32R siRNA treated worms; control worms took a further 2 h to complete migration . When compared , the time course migration pattern of Gp-flp-32 and Gp-flp-32R silenced worms was very similar; there was no statistically significant difference between the migration of Gp-flp-32 and Gp-flp-32R siRNA treated worms over the 6 h migration period ( 2 h , 53 . 2±9 . 7% vs 57 . 1±13 . 9% migration respectively , P>0 . 05; 4 h , 84 . 3±6 . 3% vs 90 . 9±4 . 4% migration respectively , P>0 . 05; 6 h , 100 . 0% vs 100 . 0% migration respectively , P>0 . 05; n = 3; see Fig . 4B and C ) . Together these data suggest that ( i ) Gp-flp-32R is involved in the maintenance of normal locomotory activity in G . pallida , and ( ii ) Gp-flp-32R is likely to be a FLP-32 activated GPCR , as the post-RNAi phenotypes of Gp-flp-32R and Gp-flp-32 silenced worms are closely matched . These conclusions are further supported by the localisation of Gp-flp-32 in neuronal cell bodies linked to the control of locomotion; unfortunately , repeated attempts to localise Gp-flp-32R ( ISH ) were unsuccessful and so we cannot comment on the co-localisation of peptide and receptor . Heterologously expressed C26F1 . 6 is activated by two peptides; FLP-7 ( TPMQRSSMVRFamide ) and FLP-32 ( AMRNALVRFamide; [13] ) . Both of these peptides display a structurally conserved C-terminus in C . elegans and many other nematode species ( VRFamide ) . However , in G . pallida , flp-7 encodes three peptides with an alternative ARFamide C-terminal motif [11] . Functional characterisation of the C . elegans VRFa receptor 1 ( C26F1 . 6 ) identified SMVRFamide as the most active truncated form of the FLP-7 peptide , suggesting that the terminal five amino acids are the most important for receptor activation [13] . This raises a question as to whether the APLDRSA ( M/L/I ) ARFamide peptides encoded by G . pallida flp-7 , which do not fit the VRFamide C-terminal model for C26F1 . 6 activation , would activate the VRFa receptor 1 homologue ( Gp-flp-32R ) in G . pallida . Preliminary data from G . pallida flp-7 RNAi experiments suggest that flp-7 peptide products do not interact with Gp-FLP-32R , as post-RNAi phenotypes for Gp-flp-7 and Gp-flp-32R silenced worms do not match . Further , G . pallida flp-7 localisation patterns are distinct from those reported for Gp-flp-32 ( unpublished data ) . Based on these observations it seems reasonable to propose that Gp-FLP-32 ( AMRNALVRFamide ) could be the primary , but not necessarily the sole ligand for Gp-FLP-32R . Curiously , C . elegans RNAi screens have revealed that VRFa receptor 1 ( C26F1 . 6 ) silencing is characterised by hyperactive egg laying activity , caused by the blunted inhibitory activity of VRFa receptor 1 ligands in the neuronal circuitry surrounding the reproductive apparatus [26] . However , this is contradictory to evidence that flps which encode the most potently active VRFa receptor 1 ligands ( TPMQRSSMVRFamide and AMRNALVRFamide; [13] ) are localised throughout the nematode nervous system , and are not limited to neurons associated with the reproductive apparatus [18] . Nevertheless , the findings described here clearly demonstrate that the Gp-flp-32R has significant importance in the modulation of motor functions in G . pallida that include , but are not necessarily limited to , the control of normal locomotory activity . Here we report the characterisation and interrogation of FLP-32/FLP-32R function in the PCN , G . pallida . The data indicate that this ligand-receptor pair may interact in PPNs to depress normal locomotory activities , modulating migration and plant root infection behaviours . This is the first functional characterisation and putative deorphanisation of a neuropeptide receptor in a parasitic nematode using reverse genetic tools , and fosters the validation of novel neuropeptidergic control targets . These data support the potential candidature of the FLP-32R as a target for agonistic drugs that would slow and potentially disrupt normal parasite locomotory behaviours .
Globodera pallida ( Pa2/3 ) were collected from potato plants of the Cara cultivar and maintained at the Agri-Food and Bioscience Institute ( AFBI ) , Belfast , Northern Ireland . Pre-parasitic J2s were hatched from cysts in fresh potato root diffusate at 15°C in complete darkness . Freshly hatched J2s were washed briefly in DEPC-treated spring water and used immediately in experiments . The roots of greenhouse maintained , susceptible tomato species infected with M . incognita were harvested and washed rigorously in water . Egg masses were removed by brief dissolution in sodium hypochlorite ( 2 . 5% v/v ) and free eggs washed thoroughly in water . Eggs were isolated by sequential washing through a nested sieve series and placed in tomato root diffusate under complete darkness at room temperature . Freshly hatched pre-parasitic J2 worms were recovered and used immediately in experiments . Basic Local Alignment Search Tool ( BLAST ) searches of the G . pallida genome were performed using the Wellcome Trust Sanger Institute BLAST server at http://www . sanger . ac . uk/cgi-bin/blast/submitblast/g_pallida . BLAST searches for a putative G . pallida VRFamide/FLP-32 receptor ( Gp-flp-32R ) candidate were conducted using the C . elegans VRFamide activated receptor 1 ( C26F1 . 6; [13] ) as a query in a translated nucleotide BLAST ( tBLASTn ) search of the G . pallida November 2010 supercontig assembly with an expect value of 100 . The high scoring return sequences in each hit were combined and translated into a single amino acid sequence in six reading frames ( http://web . expasy . org/translate/ ) , the correct reading frame was then subjected to reciprocal tBLASTn and protein BLAST ( BLASTp ) searches of the C . elegans non-redundant nucleotide ( nr/nt ) database at the National Centre for Biotechnology Information BLAST server ( http://blast . ncbi . nlm . nih . gov/ ) , using default settings . The identity of the top scoring reciprocal BLAST hit was regarded as the identity of the original G . pallida hit . Finally , putative Gp-flp-32R hits were analysed for identity and transmembrane domain structure using the online InterProScan ( http://www . ebi . ac . uk/Tools/pfa/iprscan/ ) , TMpred ( http://www . ch . embnet . org/software/TMPRED_form . html ) and HMMTOP ( http://www . enzim . hu/hmmtop/ ) servers . All BLAST searches were performed between April and October 2011 . G . pallida flp-32 ( GenBank accession number CV578361 ) and M . incognita flp-32 ( GenBank accession number CN443314 ) were previously identified through interrogation of ESTs [10] , [11] . Messenger RNA was extracted from approximately 300 G . pallida J2s using Dynabeads mRNA Direct kit ( Life Technologies ) according to the manufacturer's instructions . Separate populations of 5′ and 3′ Rapid Amplification of cDNA Ends ( RACE ) -ready cDNA were generated using the SMARTer RACE cDNA Amplification kit ( Clontech ) , as described by the manufacturer's instructions . Gene specific primers ( GSP ) were designed against a putatively assigned G . pallida flp-32 EST ( Gp-flp-32; GenBank accession number CV578361; [11] ) and a putative G . pallida flp-32R genome hit ( GenBank accession number JQ685132; see Table 1 ) , and were used in 50 µl PCR reactions to confirm the expression presence of the gene transcript of interest as follows: 5 µl 10× PCR buffer ( Life Technologies ) , 3 µl MgCl2 ( 50 mM , Life Technologies ) , 1 µl dNTP mix ( 10 mM , Promega ) , 1 µl of each sense and antisense GSP primer ( 20 µM ) , 1 µl cDNA template , 0 . 3 µl Platinum Taq DNA Polymerase ( 5 U/µl , Life Technologies ) , ddH2O to 50 µl . Thermal cycling conditions were as follows: initial denaturation and ‘hot start’ at 94°C for 2 min , followed by 40 cycles of 94°C 1 min , 55°C 1 min , and 72°C for 1 min , and a final extension step of 72°C for 7 min . Gp-flp-32 and Gp-flp-32R RACE GSPs ( RGSP , see Table 2 ) were used in 3′ RACE and 5′ RACE reactions . The components of 50 µl RACE PCR reactions were as follows: 5 µl 10× PCR buffer ( Life Technologies ) , 3 µl MgCl2 ( 50 mM , Life Technologies ) , 1 µl dNTP mix ( 10 mM , Promega ) , 2 . 5 µl RACE-ready 5′ or 3′ cDNA template , 5 µl 10× Universal Primer Mix ( UPM ) , 1 µl sense or antisense RGSP ( 20 µM ) , 0 . 3 µl Platinum Taq DNA Polymerase ( 5 U/µl , Life Technologies ) , ddH2O to 50 µl . RACE PCR reactions were carried out using the thermal cycling conditions above with annealing temperatures of 60–65°C . Finally , GSPs were designed to confirm the open reading frame ( ORF ) of both transcripts ( see Table 1 . ) . All PCR reaction products were viewed on a 1% agarose/Tris acetate EDTA ( TAE ) gel containing 0 . 0075% ( v/v ) ethidium bromide ( 10 mg/ml ) , and products of the appropriate size were PCR cleaned using Charge Switch PCR Clean-up kit ( Life Technologies ) or gel purified with Purelink Quick Gel Extraction kit ( Life Technologies ) . Products were cloned into the pCR 2 . 1 TOPO vector in One Shot Chemically Competent TOP10 Escherichia coli ( Life Technologies ) . At least three individual clones ( per PCR product ) were sequence verified by GATC Biotech ( http://www . gatc-biotech . com ) . Return sequences were analysed using Vector NTI Advance Alignx ( Life Technologies ) . siRNAs were designed against each target transcript ( Gp-flp-32 , GenBank accession number JQ685131; Gp-flp-32R , GenBank accession number JQ685132; Mi-flp-32 , GenBank accession number CN443314 ) , a species specific positive control ( Gp-ace , GenBank accession number FJ499505 ) , and a non-native negative control derived from the free-living flatworm Macrostomum lignano ( GenBank accession number EG956133; see Table 2 . ) . siRNAs were synthesised using the Silencer siRNA Construction Kit ( Ambion , supplied by Life Technologies ) according to the manufacturer's instructions , eluted in DEPC treated spring water and stored in 10 µl aliquots at −80°C until use . Approximately 500 G . pallida ( or M . incognita ) J2s were soaked in 0 . 1 mg/ml target siRNA ( Gp-flp-32 or Gp-flp-32R or Mi-flp-32 ) , non-native negative control siRNA , or DEPC treated spring water ( untreated negative control ) . All siRNAs were diluted to a final volume of 50 µl in DEPC treated spring water , soaks were carried out in triplicate , and in RNase-free hydrophobically lined microcentrifuge tubes for 24 h at 15°C in complete darkness . Following the 24 h soaking period worms were washed three times in DEPC treated spring water , transferred to a flat bottom microcentrifuge tube for mRNA extraction , or phenotypically assessed in post-RNAi migration or infection assays . Following siRNA soaking experiments G . pallida mRNA was extracted with Dynabeads mRNA Direct kit ( Life Technologies ) , treated with DNase I ( Ambion TURBO DNase , Life Technologies ) , and used as a template for cDNA synthesis using the Applied Biosystems High Capacity RNA-to-cDNA reverse transcription kit ( Life Technologies ) according to manufacturer's instructions . To assess transcript knockdown , target and housekeeping reference gene transcripts were amplified from each cDNA in triplicate qPCR's using FastStart SYBR Green Master ( Roche ) . All qPCR primers ( Table 1 . ) were designed using Primer3Plus software ( http://www . primer3plus . com/ ) and optimised for working concentration and annealing temperature prior to use with a RotorGene Q 5-plex HRM qPCR instrument ( Qiagen ) . The efficiency of each PCR reaction was calculated using Real-time PCR Miner ( http://www . miner . ewindup . info/Version2; [27] ) and used in the relative quantification of target gene transcripts by the augmented comparative Ct method ( ΔΔCt; [28] ) . Changes in target gene transcript abundance were analysed by one-way ANOVA and Tukey's Honestly Significant Difference ( HSD ) post-test , using GraphPad PRISM Version 5 package for Windows ( GraphPad Software , Inc . ) . Data with probabilities of less than 5% ( P<0 . 05 ) were deemed statistically significant . In situ hybridisation ( ISH ) probe templates ( 201–232 bp ) were generated by PCR ( cycling conditions and reaction mixtures described above ) using GSPs ( see Table 1 . ) against the target transcripts Gp-flp-32 and Gp-flp-32R , and a positive control ( Gp-flp-12; [17] ) . PCR products were viewed and sequence verified as described above . Digoxigenin ( DIG ) -labelled single stranded DNA ( ssDNA ) probes with sense and antisense polarity were generated from cDNA templates by the LATE-PCR method [30] , in the following reaction: 5 µl 10× PCR buffer ( Life Technologies ) , 3 µl MgCl2 ( 50 mM , Life Technologies ) , 2 µl DIG dNTP mix ( Roche ) , 1 µl each of sense and antisense primers ( 20 µM or 1 µM according to polarity of probe ) , 2 µl corresponding ISH probe template , 0 . 3 µl Platinum Taq DNA Polymerase ( 5 U/µl , Life Technologies ) , and ddH2O to 50 µl . ISH was carried out according to methods previously described [17] . Hybridised probes were detected with substrate ( 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium tablet; BCIP/NBT , Sigma-Aldrich ) for up to 2 h at room temperature . Specimens were mounted on glass slides and photographed using a Leica DFC300FX camera and Leica FW4000 V 1 . 2 software with a Leica DMR light microscope . A polyclonal antiserum was raised to the single peptide encoded by flp-32 ( anti-AMRNALVRFamide ) in guinea pig ( Genosphere Biotechnologies , France ) , N-terminally coupled to KLH and affinity purified . Approximately 1000 freshly hatched G . pallida J2s were immunostained using the indirect immunofluorescence technique [31] , using methods previously described [32] . Anti-AMRNALVRFamide primary antiserum was used at 1/100 working dilution , and worms were viewed on a Leica AOBS SP2 confocal scanning laser microscope . Controls included the omission of primary antiserum , replacement of primary antiserum with pre-immune serum from the donor species , and pre-adsorption of the primary antiserum with ≥250 ng of AMRNALVRFamide and an additional FLP peptide ( NGAPQPFVRFamide ) . Pre-adsorption in NGAPQPFVRFamide did not alter staining patterns observed . | Plant pathogenic nematodes compromise plant health and productivity globally and are an increasing problem due to the lack of efficient control measures . The nematode nervous system depends heavily on small proteins ( neuropeptides ) for communication between nerve cells and other nerve cells or other cell types . The disruption of neuropeptide signalling would dysregulate normal behaviour , offering an attractive approach to parasite control . One major group of nematode neuropeptides are the FMRFamide-like peptides ( FLPs ) that alter nematode behaviour by acting on receptors designated G-protein coupled receptors ( GPCRs ) . GPCRs are attractive targets based on their potential ‘druggability;″ indeed they are targets for many human medicines . This study investigates the functional biology of flp-32 , a commonly expressed nematode flp , and a novel FLP-32 receptor in a plant pathogenic nematode of major agricultural importance , Globodera pallida . We show that FLP-32 occurs widely in these parasites and interacts with a novel FLP-32 receptor to modulate their behaviour , affecting their movement and the rate at which they infect host plants . These data indicate that chemicals that activate the FLP-32 receptor in these parasites could effectively slow the worms , potentially making them less successful parasites . The conservation of the FLP-32 ligand and receptor across many different nematode parasites adds to its appeal as a potential target for broad-spectrum parasite control . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"neuroscience",
"rna",
"interference",
"small",
"molecules",
"neuroscience",
"gene",
"function",
"immunochemistry",
"molecular",
"genetics",
"signaling",
"pathways",
"gene",
"expression",
"biology",
"molecular",
"biology",
"drug",
"discovery",
"biochemistry",
... | 2013 | flp-32 Ligand/Receptor Silencing Phenocopy Faster Plant Pathogenic Nematodes |
Human African trypanosomiasis ( HAT ) , or sleeping sickness , results from infection with the protozoan parasites Trypanosoma brucei ( T . b . ) gambiense or T . b . rhodesiense and is invariably fatal if untreated . There are 60 million people at risk from the disease throughout sub-Saharan Africa . The infection progresses from the haemolymphatic stage where parasites invade the blood , lymphatics and peripheral organs , to the late encephalitic stage where they enter the central nervous system ( CNS ) to cause serious neurological disease . The trivalent arsenical drug melarsoprol ( Arsobal ) is the only currently available treatment for CNS-stage T . b . rhodesiense infection . However , it must be administered intravenously due to the presence of propylene glycol solvent and is associated with numerous adverse reactions . A severe post-treatment reactive encephalopathy occurs in about 10% of treated patients , half of whom die . Thus melarsoprol kills 5% of all patients receiving it . Cyclodextrins have been used to improve the solubility and reduce the toxicity of a wide variety of drugs . We therefore investigated two melarsoprol cyclodextrin inclusion complexes; melarsoprol hydroxypropyl-β-cyclodextrin and melarsoprol randomly-methylated-β-cyclodextrin . We found that these compounds retain trypanocidal properties in vitro and cure CNS-stage murine infections when delivered orally , once per day for 7-days , at a dosage of 0 . 05 mmol/kg . No overt signs of toxicity were detected . Parasite load within the brain was rapidly reduced following treatment onset and magnetic resonance imaging showed restoration of normal blood-brain barrier integrity on completion of chemotherapy . These findings strongly suggest that complexed melarsoprol could be employed as an oral treatment for CNS-stage HAT , delivering considerable improvements over current parenteral chemotherapy .
Human African trypanosomiasis ( HAT ) , also known as sleeping sickness , is endemic in 36 countries , in sub-Saharan Africa where 60 million people are at risk from infection [1] , [2] . The disease is caused by the protozoan parasites Trypanosoma brucei ( T . b . ) gambiense in West Africa and T . b . rhodesiense in East Africa and is spread by the bite of the tsetse fly vector [1] , [2] . Infection with T . b . gambiense usually results in a disease that follows a chronic course which can last for up to several years before death ensues while T . b . rhodesiense infection follows an acute pattern with death occurring in only weeks to months [3] . In both infections the disease progresses in two stages , the early or haemolymphatic stage and the late encephalitic or CNS-stage [3] . During the early-stage the parasites migrate from the site of the tsetse fly bite and spread throughout the body via the blood and lymph , invading the peripheral organs . The trypanosomes then cross the blood-brain barrier ( BBB ) and migrate into the CNS to cause the characteristic clinical manifestations of CNS-stage disease such as alteration of sleep patterns , neuropsychiatric symptoms and a variety of motor and sensory disturbances [4] . If the disease is diagnosed during the early stage it can be treated with pentamidine ( for T . b . gambiense ) or suramin ( for T . b . rhodesiense ) [5] . If the infection has reached the CNS , T . b . gambiense infections can be treated with either a concise 10-day regimen of melarsoprol [6] , [7] or the recently developed nifurtimox-eflornithine combination therapy ( NECT ) [8] . In the case of CNS-stage T . b . rhodesiense infections the only treatment option currently available is a lengthy melarsoprol schedule comprising 3–4 cycles of a series of 3–4 injections , of increasing melarsoprol concentration , separated by a 7–10 day interval between each cycle [6] , [9] . Melarsoprol ( Figure 1A ) is a highly lipophilic molecule that is poorly soluble in water with a log POW of 2 . 53 and a solubility of only 6 mg/L at 25°C [10] . Despite these properties the drug is a potent trypanocide and has been used for the treatment of HAT since its introduction in 1949 [11] . The limited solubility of melarsoprol precludes its oral delivery as only a small fraction of the drug is absorbed through the gastrointestinal tract . Currently melarsoprol is produced as a 3 . 6% solution in propylene glycol ( Arsobal ® ) which restricts its administration to the intravenous route . The treatment schedules employed are protracted , excruciatingly painful and require continuous hospitalization . In addition , treatment with Arsobal® is associated with numerous adverse effects including severe tissue necrosis at the injection site , neuropathy , and gastrointestinal upset [4] . However , the most serious adverse reaction is the development of a post-treatment reactive encephalopathy ( PTRE ) which occurs in 10% of all treated patients , 50% of whom die as a result . Arsobal® treatment is therefore responsible for the death of 5% of all patients given the drug [1] , [12] . Although the pathogenesis of the PTRE remains unclear , several hypotheses have been postulated to explain its occurrence . These include direct arsenical toxicity [13] , [14] , autoimmune reactions [15] and pro-inflammatory immune response directed against trypanosomes remaining within the CNS following systemic clearance of the parasites [16] , [17] or parasite antigen released as a consequence of chemotherapy [18] . The severity of the complications associated with Arsobal® chemotherapy make accurate staging of the disease via cerebrospinal fluid analysis absolutely essential both to manage proven CNS-stage infections appropriately and to prevent unnecessary administration of this highly toxic drug to early stage patients [19] . Cyclodextrins are naturally occurring cyclic oligosaccharide molecules composed of six or more glycopyranose units linked by α-1 , 4 gycosidic bonds . They take the form of a truncated cone or torus with a hydrophilic exterior and a hydrophobic interior cavity which can be occupied by various guest molecules [20] . Cyclodextrins have been widely utilized by the pharmaceutical industry to alter the physiochemical properties of a variety of drugs through enhancing their solubility and oral bioavailability and decreasing their toxicity [10] , [21] . Complexation of melarsoprol with either hydroxypropyl-β-cyclodextrin ( mel/HPβCD ) or randomly methylated-β-cyclodextrin ( mel/RAMβCD ) ( Figure 1A ) has been shown to increase the inherent solubility of the drug by a factor of 7 . 2×103 , which raises the possibility that the melarsoprol cyclodextrin complexes could be efficacious when delivered via the oral route for the treatment of trypanosomiasis [10] . In the current study the efficacy of the melarsoprol cyclodextrin inclusion complexes was investigated using both in vitro and in vivo methodologies and compared with that of contemporary melarsoprol formulations . The effect of oral drug treatment on the BBB was examined using MRI , and both the CNS parasite load and the CNS neuroinflammatory response monitored throughout the treatment regimen . We show here that melarsoprol cyclodextrin complexes are orally effective and non-toxic in curing CNS-stage trypanosome infections in mice .
Trypanotoxicity was determined using an adapted version of the Alamar Blue assay [22] . Bloodstream form T . brucei brucei ( strain 427 ) were cultivated in HMI-9 medium ( BioSera Ltd . , UK ) [23] supplemented with 2 mM β-mercaptoethanol ( Sigma-Aldrich , UK ) and 10% fetal calf serum ( BioSera Ltd . , UK ) at 37°C in a humidified 5% CO2 environment . Parasites ( 100 µl of 1×104 trypanosomes/ml ) were added to wells of 96-well plates containing doubling dilutions of the drugs ( 100 µl ) ranging in final concentration from 100 µM to 24 pM and incubated for 48 hours . Alamar Blue reagent ( 20 µl , 0 . 49 mM in PBS , pH 7 . 4; Sigma-Aldrich , UK ) was added to each well and , after 24 hours , the fluorescence was measured using a LS 55 luminescence spectrophotometer ( PerkinElmer Life and Analytical Sciences , USA ) set at excitation and emission wavelengths of 530 nm and 590 nm respectively . Data was analysed and inhibitory concentration ( IC50 ) values determined with Prism 5 . 0 ( GraphPad Software , USA ) software . The experiment was performed in duplicate on three independent occasions . A well established and characterised model of CNS-stage human African trypanosomiasis was employed throughout this investigation . Briefly , female CD-1 mice ( Charles River Laboratories ) ( 20–30 g body weight ) were infected with 3×104 Trypanosoma brucei brucei GVR35 parasites by intraperitoneal injection . The infection was allowed to progress until day 21 without drug intervention . At this point the parasites have established within the CNS and the infection has entered the encephalitic stage . Infection was confirmed in all mice prior to drug treatment by examination of a wet blood film for the presence of parasites . To determine whether a treatment regimen was curative , blood smears were examined for the presence of parasites on a weekly basis for a period of 60 days . If the animals relapsed to parasitaemia the regimen was considered unsuccessful and the mice were killed . Mice that remained aparasitaemic for the duration of the monitoring period were killed , the brains excised and lightly homogenised in PBS supplemented with 1 . 5% glucose w/v and injected intraperitoneally into a clean recipient animal . This mouse was then monitored for the presence of parasites for a further 60 days . If the mouse remained aparasitaemic the treatment regimen was considered successful . All animal procedures were authorised under the Animals ( Scientific Procedures ) Act 1986 and approved by the University of Glasgow Ethical review Committee . Trypanosome load within the brain was determined by real-time quantitative PCR . Briefly , whole brains were homogenised and digested with proteinase K ( AppliChem GmbH ) and DNA extracted from a 25 mg sample of the homogenate ( Qiagen , DNeasy Tissue kit ) . The concentration of the extracted DNA was assessed by measuring the absorbance and the sample diluted to 20 ng/ml . The reaction mix was comprised of; Taqman Brilliant II master mix ( Agilent ) , 0 . 05 pmol/µL of each primer , 0 . 1 pmol/µL probe ( Eurofins MWG Operon ) and 100 ng DNA template . The amplification was performed on a MxPro 3005 thermocycler ( Agilent ) . The mel/HPβCD and mel/RAMβCD inclusion complexes were prepared as previously described [10] . Each complex was dissolved in sterile water and administered at dose rates of 0 . 0125 , 0 . 025 , 0 . 05 , 0 . 1 and 0 . 2 mmol/kg ( equivalent to 4 . 975 , 9 . 95 , 19 . 9 , 39 . 8 , and 79 . 6 mg/kg ) of the active ingredient , melarsoprol . Non-complexed HPβCD and RAMβCD ( Sigma ) were used as control treatments and administered at dose rates equivalent to 0 . 1 mmol/kg of the complexed agent . Contemporary melarsoprol and the melaminophenyl arsine derivatives [24] , melarsamine hydrochloride ( MelCy ) ( Cymelarsan® ) and melarsonyl potassium ( MelW ) ( Trimelarsen® ) were prepared as solutions or fine suspensions in sterile water and administered at a dose of 0 . 05 mmol/kg . All drug treatments were delivered orally by gavage , once per day for a period of 7 days beginning on day 21 post-infection . Body weights were measured in groups of uninfected mice before and after completion of treatment and clinical appearance was monitored using an established visual assessment scale [25] throughout the drug regimens to assess overt signs of drug toxicity . MRI was performed on two mice at day 21 post-infection prior to drug treatment . The mice were re-scanned at 24 hours , 8 and 15 days after completion of chemotherapy . Uninfected mice ( n = 3 ) were also examined . All scans were performed as described previously [26] . Briefly , mice were anaesthetised and the tail vein was cannulated with a 26 gauge×19 mm cannula to facilitate contrast agent administration during MRI scanning . The animal was placed into a mouse cradle and restrained using ear and tooth bars to minimise head movement . Anaesthesia was maintained throughout the procedures and respiration , heart rate and body temperature were observed . The animal was maintained normothermic by an enclosed warm water circuit . MRI was performed on a Bruker Biospec 7 T/30 cm system equipped with an inserted gradient coil ( 121 mm ID , 400 mT/m ) and a 72 mm birdcage resonator . A surface coil was used for brain imaging . The scanning protocol consisted of a RARE T2 weighted scan [effective TE ( echo time ) 76 ms , TR ( repetition time ) 5362 ms , 25 averages , matrix 176×176 , FOV ( field of view ) 17 . 6×17 . 6 mm , 20 contiguous coronal slices of 0 . 4 mm thickness] followed by a RARE T1 weighted scan ( effective TE 9 ms , TR 8000 ms , 20 averages , matrix 176×176 , FOV 17 . 6×17 . 6 mm , 20 contiguous coronal slices of 0 . 4 mm thickness ) . Following the RARE T1 weighted scan 0 . 1 ml of a solution containing 50 µL gadolinium-diethylenetriamine penta-acetic acid ( Gd-DPTA Magnevist®; Bayer ) and 50 µL of sterile water was injected via the tail vein cannula . Five minutes later the T1 weighted scan was repeated . Gd-DTPA cannot readily cross the intact blood brain barrier due to its charge and high molecular weight [27] . Extravasation of Gd-DTPA observed within the parenchyma demonstrates an impairment of the BBB integrity . Images were analysed using Image J software ( http://rsbweb . nih . gov/ij/ ) . Contrast enhancement maps were generated from the the per and post-contrast T1 weighted scans according to the equation: Enh = ( Spost−Spre ) ÷Spre where Spost = post contrast agent signal and Spre = pre-contrast agent signal . Regions of interest ( ROIs ) were manually defined to include the entire brain and meninges . The mean percentage signal change for each brain slice was then calculated and signal enhancement maps generated . Following sacrifice the brains were excised and fixed in 4% neutral buffered formalin , paraffin wax blocks prepared and sections of 3 µm thickness cut and stained with haematoxylin and eosin . These sections were examined by two independent assessors and the severity of the neuropathological reaction graded on a scale of 0–4 where 0 represented normal pathology with no indications of inflammation and a grade of 4 was characterised by the presence of a severe meningoencephalitis with the presence of inflammatory cells in the brain parenchyma [26] , [28] ( Table S5 ) . Immunocytochemistry was performed to detected T-cells ( rabbit anti-CD3 ) , B-cells ( rat anti-B220 ) and macrophages ( rat anti-F4/80 ) following a standard peroxidise anti-peroxidase protocol using the Dako® EnVision system and DAB visualisation . Data were analyzed using analysis of variance methods and the General Linear Model ( GLM ) procedure in Minitab Version 16 followed by multiple pair wise comparison tests . This identified significant main effect differences between groups of uninfected animals , infected animals and treated animals . In studies with measurements over time the GLM procedure provided tests for treatment and time effects and their interaction . Proportions of mice relapsing in different treatment groups were compared using a Tukey-type multiple comparison test for proportions [29] . Changes in body weight were investigated using a paired t-test . P values of less than 5% were considered to be statistically significant . Where appropriate data were log transformed prior to analysis . Group means were plotted showing means and their standard errors , and the size of treatment effects were estimated using differences between group means and their 95% confidence intervals . Log dose response curves provided estimates of IC50 concentrations .
To determine whether the complexed melarsoprol retains its trypanocidal properties a modification of the Alamar blue assay [22] was used to investigate the inhibitory concentration ( IC50 ) of the complexed melarsoprol molecules in comparison to standard melarsoprol and an unrelated trypanocidal drug , diminazene aceturate , in an in vitro trypanosome culture system . The IC50 values determined for mel/HPβCD and mel/RAMβCD were 21 . 6 nM and 8 . 8 nM respectively ( Figure 1B ) . Standard melarsoprol returned an IC50 value of 6 . 9 nM . Statistical analyses of the Alamar blue data revealed no significant changes in the trypanocidal activity of melarsoprol following complexation when compared to the standard form of the drug ( P = 0 . 2002 , P = 0 . 9999; mel/HPβCD and mel/RAMβCD respectively ) . The HPβCD and RAMβCD molecules alone did not display any trypanocidal activity ( Table S1 , Figure 1B ) . The ability of the complexed melarsoprol compounds to cure CNS-stage trypanosome infections was investigated in a well established and characterized murine model of the disease . The drugs were administered by oral gavage each day at doses ranging from 0 . 0125 mmol/kg to 0 . 2 mmol/kg for a seven day period . All animals became aparasitaemic following the melarsoprol regimens; however , all mice treated at the 0 . 0125 mmol/kg level relapsed to parasitaemia . A relapse to parasitaemia was also detected in one third of the mice treated with mel/HPβCD and one sixth of the mice given mel/RAMβCD at the 0 . 025 mmol/kg level . Successful cures were obtained in all mice treated with the 0 . 05 mmol/kg , 0 . 1 mmol/kg or 0 . 2 mmol/kg dosage of either complex , indicating that 0 . 05 mmol/kg was the minimum dosage necessary to achieve successful cures . Animals given HPβCD or RAMβCD alone remained parasitaemic throughout the procedure ( Figure 1C ) . Paired t-test analysis detected no evidence of decreased body weight in uninfected mice following 7-days of oral drug treatment . A significant ( P = 0 . 019 , 95% confidence interval 0 . 213 g , 1 . 954 g ) increase was detected between the mean body weight of the pre- and post treatment groups ( 25 . 83±0 . 696 g; 26 . 92±0 . 890 g respectively ) . No adverse clinical signs were detected at any point during the chemotherapy regimen with the mice remaining alert and showing good coat condition . The efficacy of melarsoprol ( MelB ) and the water soluble melaminophenyl arsine derivatives [24] , melarsamine hydrochloride ( MelCy ) and melarsonyl potassium ( MelW ) ( Figure 1A ) when administered per os at a dose of 0 . 05 mmol/kg for seven consecutive days , during CNS-stage infections was investigated . No cures were obtained in the mice treated with MelCy or MelW and only 33% of the mice given MelB were successfully cured ( Figure 1D ) . Taqman real-time PCR was performed ( Figure S 1 ) to determine the parasite numbers present within the brain tissue prior to chemotherapy and at 24 hours after each oral dose of mel/HPβCD or mel/RAMβCD ( Figure 2A ) . Animals killed on day 21 post-infection , prior to receiving any drug treatment showed a mean parasite load of 626±82 . 8 ( mean ± SE ) . Following a single dose of mel/HPβCD or mel/RAMβCD the parasite numbers detected within the brain were significantly ( P<0 . 001 ) reduced ( 68 . 1±14 . 7; 66 . 2±10 . 8 respectively ) . The decrease in parasite numbers continued in a stepwise manner with successive treatments until the trypanosomes were completely cleared from the brain ( Figure 2B & C , Table S2 & S3 ) . Interaction plots comparing the mean copy numbers detected after each dose of mel/HPβCD and mel/RAMβCD show that there are no significant differences between the clearance rates achieved by either of the drugs ( Figure 2D ) . From the Taqman results it is apparent that both forms of complexed melarsoprol clear the trypanosomes from the brain in a rapid and efficient manner with a reduction of greater than 80% of the parasite load 24 hours after the initial drug treatment . We determined the effect of oral treatment with mel/HPβCD on BBB function using MRI . Mice were examined prior to treatment , and 24 hours , 8 and 15 days following the chemotherapy regimen ( Figure 3A ) . MRI scans were performed before and after the injection of Magnevist® contrast agent ( Gd-DPTA ) [27] and signal enhancement maps generated as previously described [26] . Changes in BBB integrity were investigated in two infected mice scanned at day 21 post-infection and compared with scans prepared in the same animals 24 hours , 8 days and 15 days after completing a 7 day oral course of mel/HPβCD as well as those from uninfected mice ( n = 3 ) . At day 21 post-infection the BBB was significantly compromised ( 17 . 87±1 . 62 ) ( Figure 3B , Figure 4 , Table S4 ) . Signal enhancement was present throughout the brain with highest signal change found in the ventricular region . Changes in signal intensity were also apparent in the cerebral cortex , hypothalamus , hippocampus and median eminence ( Figure 4 ) . However , by 24 hours after completion of the chemotherapy ( Figure 4 ) the percentage signal change ( 7 . 93±0 . 455 ) had dropped significantly ( P<0 . 0001 ) and was comparable ( P = 0 . 9296 ) to that seen in uninfected mice ( 7 . 11±0 . 162 ) ( Figure 3B , Figure 4 ) indicating that by this point the integrity of the BBB had become re-established . The integrity of the barrier was maintained in all subsequent scans performed at 8 days ( 9 . 25±0 . 596 ) ( Figure 3B , Figure 4 ) and 15 days ( 6 . 55±0 . 463 ) ( Figure 3B , Figure 4 ) after completion of the treatment schedule ( Table S4 ) . The severity of the neuropathological response to the trypanosome infection and drug treatment was determined in mice killed 15 days after completing the treatment schedule and compared to animals killed at day 21 post-infection prior to receiving chemotherapy using a well established grading scale [28] ( Table S5 ) . Pathological examination of the brains prepared from animals prior to drug treatment showed mild neuroinflammatory changes ( 1 . 5±0 . 158 ) with the presence of an inflammatory cell infiltrate in the meninges and Virchow–Robin spaces ( Figure 5 ) . Some perivascular cuffs were also apparent surrounding the blood vessels in the hippocampus ( Figure 5 ) . The cellular infiltrate was composed mainly of lymphocytes , and macrophages ( Figure 6 ) . A significant ( P = 0 . 0366 ) resolution of this neuroinflammation ( 1 . 083±0 . 083 ) was apparent in mice killed 15 days after completion of the oral mel/HPβCD regimen . This represents a mean decrease of 27 . 8% with a 95% confidence interval ( 0 . 032 , 0 . 801 ) . Only a few inflammatory cells could be detected in the meninges of these animals accompanied by very mild perivascular infiltration of the occasional blood vessel in the hippocampus ( Figure 5 & 6 ) .
New drugs to treat HAT remain an urgent priority . In spite of some recent positive developments [30] the situation remains precarious as evidenced by the failure , late in clinical trials , of pafuramidine ( DB289 ) . Ideally new drugs should be orally available and of equal or better efficacy than current drugs with improved safety . Melarsoprol is the only drug suitable to treat CNS-stage rhodesiense disease and remains in use in some areas for gambiense . Its use , however , is tainted by its being administered parenterally and through its well documented adverse events . The study reported here shows that complexation of melarsoprol with the cyclodextrin molecules does not affect the trypanocidal properties of the compound and appreciably enhances the ability of the drug to cure CNS-stage trypanosome infections when given as an oral chemotherapy regimen . The improved oral bioavailability seen in these investigations is most likely due to the increased solubility and reduced degradation kinetics of the drug following complexation with the cyclodextrin molecules [10] , [31] . Further , cyclodextrins can also act as carriers , delivering the drug directly to the intestinal membrane while protected within the cavity of the molecule [32] , [33] . Consistent with our findings is the improved oral bioavailability reported with a variety of cyclodextrin drug inclusion complexes including anti-parasitic agents . The potent anti-malarial drug artemisinin has low aqueous solubility that severely limits its absorption following oral administration . Complexation of artemisinin with cyclodextrin molecules has been shown to improve the pharmacokinetic profile of the drug compared with Artemisinin 250® when given per os [34] . This has also been demonstrated for the anti-helminthic drug albendazole [35] . The pathogenesis of the PTRE associated with standard melarsoprol treatment is currently unknown although several hypotheses have been suggested [13]–[18] . However , it is probably caused , at least in part , by an acute toxic reaction to low levels of arsenic within the CNS following delivery of an intravenous bolus of the arsenical drug [13] , [14] . On the basis of our combined data the lack of toxicity and the resolution of the CNS inflammatory reaction shown following oral treatment with melarsoprol cyclodextrin inclusion complexes can most easily be explained by the more controlled delivery of the trypanocidal drug to the brain following a sustained absorption from the gut compared with that of an intravenous bolus . Consistent with this explanation are the extremely low levels of arsenic in the brain tissue following chemotherapy which were below the level of detection ( <5 ng/mL ) of the GC-MS assay employed ( unpublished data ) and the extremely rapid clearance of the parasites from the CNS following drug administration . This is also reflected by the restoration of BBB integrity detected in the mice 24 hours after completion of the chemotherapy regimen . However , a direct comparison of these criteria following a curative IV regimen of Arsobal® would be required to corroborate this hypothesis . Multiple IV doses of Arsobal® cannot be administered in the murine model due to the severe venous damage caused by the propylene glycol solvent present in the drug preparation . Therefore , data regarding drug levels , parasite clearance and BBB function following IV Arsobal® remain unavailable . Taken together these findings strongly suggest that mel/HPβCD and mel/RAMβCD could be used to treat patients with CNS-stage HAT . Since these experiments were performed using a T . b . brucei model of infection it is possible that these drug complexes will not show the same activity profile when transferred to human disease . However , since the active trypanocidal component of the complex is melarsoprol , with proven effectiveness against both T . b . rhodesiense and T . b . gambiense infections , this scenario seems highly unlikely . Consequently , in the first instance , these complexes should be tested in subjects with T . b . rhodesiense , even though these comprise the minority of cases of HAT compared with T . b . gambiense , since melarsoprol is currently the only drug that can be effective in rhodesiense disease . The drugs are effective orally at dosages that could be delivered in humans . During the concise 10-day schedule for Arsobal® treatment a 60 kg patient would be given a total dose of 1320 mg of melarsoprol . In the current study , melarsoprol cyclodextrin complexes were curative when administered at 0 . 05 mmol/kg or 19 . 9 mg/kg melarsoprol daily for a seven day period . To obtain an approximate human equivalent dose ( HED ) from this data the dosage must be normalized according to body surface area which can be achieved by dividing the murine dose by a factor of 12 [36] . The HED for the complexed drugs would therefore be approximately 1 . 6 mg/kg , with a total dosage of 672 mg assuming a seven day course and a 60 kg body weight . This is a considerable reduction in the total amount of arsenical required for each drug course , even when compared with the concise schedule . This decreased arsenical dosage could also be a major factor in the apparent lack of toxicity associated with the oral regimen . These complexes rapidly clear the trypanosomes from the brain following administration , reduce the severity of the neuropathological response induced by the infection , and also restore BBB integrity following treatment . The availability of an orally administrable drug would preclude both the need for hospitalization of the patient throughout the period of treatment and the provision of highly skilled clinicians to administer the drug by slow intravenous infusion . Further , the pain and fear associated with current melarsoprol therapy would be circumvented and patients would be far more likely to be compliant in finishing the treatment course . This would have a significant positive socio-economic impact in local communities and on the already burdened health care budgets of these regions . One of the major problems in the management of HAT is that there is no clear consensus on the criteria used to classify an infection as having progressed to the CNS-stage [19] , [37] . The current WHO criteria suggest that CSF containing >5 white blood cells ( WBC ) /µL with or without the presence of trypanosomes indicates CNS-stage infection [9] . However in some T . b . gambiense infections the higher value of >20 WBC/µL has been used before commencing melarsoprol treatment [38] , [39] . This has significant implications for choosing the correct chemotherapeutic approach to best manage the infection . Inappropriate administration of melarsoprol to patients with early-stage disease exposes them to unnecessary risks form drug toxicity while failure to use melarsoprol in CNS-stage disease will have inevitably fatal consequences [19] . The use of an alternative treatment strategy without the associated adverse safety profile of the intravenous melarsoprol formulation would also obviate significantly the difficulties associated with the current methods of disease staging [19] . In conclusion , the current chemotherapy options for treatment of CNS-stage HAT are extremely limited and all involve parenteral administration of highly toxic and sometimes ineffective drugs . Moreover there are no new alternative drugs for CNS HAT likely to be used in clinical practice for at least 5–10 years [30] . Only one compound , fexinidazole , is currently in Phase I clinical trials [40] . Due to the high failure rate of novel compounds it is critical to maintain drug development in this area to ensure that effective treatments for both forms of this disease are available in the future . Sir James Black , the Nobel Laureate said ‘the most fruitful basis for the discovery of a new drug is to start with an old drug’ [30] . If melarsoprol cyclodextrin inclusion complexes , given via the oral route , prove equally efficacious in patients with HAT this would constitute one of the most significant therapeutic advances in the long history of the disease . Plans to test these drug complexes in a phase II trial in East African patients with T . b . rhodesiense are currently being formulated . | Human African trypanosomiasis ( HAT ) is caused by infection with either Trypanosoma brucei gambiense or T . b . rhodesiense and is fatal if untreated . In the late stages of the disease the parasites enter the brain , producing severe neurological symptoms . The arsenical drug melarsoprol is the only treatment available for rhodesiense disease once it has reached the brain . Melarsoprol is very poorly soluble in water which severely limits its oral bioavailability . Currently melarsoprol is solubilised in propylene glycol which restricts its administration to the intravenous route and treatment schedules are protracted and extremely painful . Further , this formulation of melarsoprol is toxic and kills 5% of all treated patients through the induction of a severe post-treatment reactive encephalopathy . In this study we show that combining melarsoprol with cyclodextrin molecules increases the oral bioavailability of the drug . In a murine model of late stage HAT oral treatment regimens employing melarsoprol cyclodextrin inclusion complexes rapidly cleared parasites from the brain , restored blood-brain barrier function and reduced the severity of the neuropathological response induced by the infection . If complexed melarsoprol is equally efficacious in patients with HAT this would not only have significant positive socio-economic impact but also constitute a major therapeutic advance in the field . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"drugs",
"and",
"devices",
"african",
"trypanosomiasis",
"parasitic",
"diseases"
] | 2011 | Melarsoprol Cyclodextrin Inclusion Complexes as Promising Oral Candidates for the Treatment of Human African Trypanosomiasis |
We report the first genome-wide association study of habitual caffeine intake . We included 47 , 341 individuals of European descent based on five population-based studies within the United States . In a meta-analysis adjusted for age , sex , smoking , and eigenvectors of population variation , two loci achieved genome-wide significance: 7p21 ( P = 2 . 4×10−19 ) , near AHR , and 15q24 ( P = 5 . 2×10−14 ) , between CYP1A1 and CYP1A2 . Both the AHR and CYP1A2 genes are biologically plausible candidates as CYP1A2 metabolizes caffeine and AHR regulates CYP1A2 .
Caffeine ( 1 , 3 , 7-trimethylxanthine ) is the most widely consumed psychoactive substance in the world with nearly 90% of adults reporting regular consumption of caffeine-containing beverages and foods [1] , [2] . Although demographic and social factors have been linked to habitual caffeine consumption , twin studies report heritability estimates between 43 and 58% for caffeine use; 77% for heavy use , and 45 , 40 , and 35% , respectively , for caffeine toxicity , tolerance and withdrawal symptoms [3] . Genetic association studies focused on candidate genes related to the pharmacokinetic and pharmacodynamic properties of caffeine have identified genes encoding cytochrome P-450 ( CYP ) 1A2 , as the primary enzyme involved in caffeine metabolism [3] , [4] . The genome-wide association approach has emerged as a powerful means for discovering novel loci related to habitual use of a second stimulant , tobacco [5] , but has not yet clearly identified genes for other common behavioral traits , including caffeine consumption . To comprehensively examine the influence of common genetic variation on habitual caffeine consumption behavior we undertook a meta-analysis of genome-wide association studies ( GWAS ) from population-based cohorts . Our study confirms the important roles of CYP1A2 and AHR in determining caffeine intake , thus supporting the utility of the GWAS approach to the discovery of loci linked to this complex behavioral trait .
We performed a meta-analysis of 47 , 341 individuals of European descent , derived from five studies within the US , the Atherosclerosis Risk in Communities ( ARIC , N = 8 , 945 ) Study , the Prostate , Lung , Colorectal , and Ovarian Cancer Screening Trial ( PLCO , N = 4 , 942 ) , the Nurses' Health Study ( NHS , N = 6 , 774 ) , the Health Professionals Follow-Up Study ( HPFS , N = 4 , 023 ) , and the Women's Genome Health Study ( WGHS , N = 22 , 658 ) . Sample characteristics are presented in Table 1 . Caffeine intake was assessed using semi-quantitative food frequency questionnaires ( FFQ ) that included questions on the consumption of caffeinated coffee , tea , soft drinks , and chocolate . Study-level genomic inflation factors ( λ ) were low ranging from 1 . 00 ( PLCO ) to 1 . 03 ( HPFS ) , suggesting that population stratification was well controlled ( Figure S1 ) . A total of 433 , 781 imputed and genotyped SNPs passed our stringent criteria for the meta-analysis . Test statistic inflation at the meta-analysis level revealed no evidence of notable underlying population substructure ( λ = 1 . 04 , Figure 1 ) . Two loci reached genome-wide significance with no evidence for significant between- study heterogeneity ( Table 2 , Figure 2 and Figure 3 , Table S1 ) . The strongest associated SNP ( rs4410790 , P = 2 . 4×10−19 , Figure S2 ) is located at 7p21 , 54 kb upstream of AHR ( aryl hydrocarbon receptor ) . The second strongest associated SNP ( rs2470893 , P = 5 . 2×10−14 , Figure S2 ) mapped to 15q24 within the bidirectional promoter of the CYP1A1-CYP1A2 locus [6] , [7] . A synonymous coding SNP ( rs2472304 , P = 2 . 5×10−7 ) in CYP1A2 exon 7 that was highly correlated with 6 other SNPs but not correlated with rs2470893 ( r2 = 0 . 18 , HapMap CEU ) was amongst the highest ranked loci in our meta-analysis ( Table 2 ) . Although we only considered variants that were imputed with high probability , we also conducted a sensitivity analysis restricting our sampling to individuals with genotyped data ( Table 2 ) . Regression coefficients remained essentially unchanged , but P-values were less significant reflecting the reduced sample size ( rs4410790: P = 4 . 0×10−18; rs2470893 P = 9 . 5×10−8 ) . Similar results were also observed when men and women were examined separately ( Table S2 ) . Had the analysis been performed instead by discovery at genome-wide significance ( P<5×10−8 ) in the WGHS followed by replication in meta-analysis of the remaining cohorts , only SNPs at the same loci would have met Bonferroni corrected standards of significance . In a post-hoc investigation of study heterogeneity in which we compared WGHS to the remaining studies combined , there was significant heterogeneity for rs4410790 ( P = 0 . 01 ) , although this could be attributable to chance . Based on the well-established biological link between smoking and AHR [8] , and CYP1A2 [9] and caffeine consumption behavior [2] , we explored the role of cigarette smoking ( Table 3 ) . Compared to our primary model that adjusted for smoking , a model not adjusted for smoking yielded slightly attenuated associations and when restricting analyses to ‘never smokers’ similar regression coefficients were observed as for the complete study population . These findings suggest that smoking is unlikely the cause of the associations observed in our GWAS of caffeine intake . We further conducted 21 candidate gene analyses and found significant gene-based associations ( Bonferroni corrected for the total number of human genes ) between CYP2C9 ( P = 0 . 023 ) , and ADORA2A ( P = 0 . 011 ) and caffeine intake in addition to CYP1A2 and AHR ( Table 4 ) .
In the first GWAS of caffeine intake in a total of 47 , 341 individuals from five U . S . studies , loci at 15q24 and 7p21 achieved genome-wide significance . CYP1A2 at 15q24 and AHR at 7p21 are attractive candidate genes for caffeine intake . At plasma concentrations typical of humans ( <100 µM ) , caffeine is predominantly ( ∼95% of a dose ) metabolized by CYP1A2 via N1- , N3- , and N7-demethylation to its three dimethylxanthines , namely , theobromine , paraxanthine , and theophylline , respectively [10] . CYP1A2 expression and activity vary 10- to 60-fold between individuals [11] . Human CYP1A2 is located immediately adjacent to CYP1A1 in reverse orientation and the two genes share a common 5′-flanking region [12] . At least 15 AHR response elements ( AHRE ) reside in this bidirectional promoter region and rs2470893 is located in AHRE6 ( originally reported as AHRE5[7] ) which correlates with transcriptional activation of both CYP1A1 and CYP1A2 [6] , [7] . CYP1A1 expression in the liver ( the target tissue for caffeine metabolism ) is low and there is little evidence that this enzyme contributes to caffeine metabolism . This contrasts with the tissue specific expression of CYP1A2 in the liver , which suggests further evidence supporting its role in caffeine metabolism . The observation that a stronger association exists for SNPs upstream of the gene suggests that variation in CYP1A2 gene expression probably affects caffeine intake . The protein product of AHR , AhR , is a ligand–activated transcription factor that , upon binding , partners with ARNT and translocates to the nucleus where it regulates the expression of a number of genes including CYP1A1 and CYP1A2 . There is marked variation in AhR binding affinity across populations , but so far no polymorphisms have been identified that account for this variation [13] . The most studied SNP , rs2066853 ( R554K ) , is located in exon 10 , a region of AHR that encodes the transactivation domain[13] . Although this SNP was associated with caffeine in the current study ( P = 0 . 0004 ) , our strongest signal mapped upstream of AHR , suggesting variation in AHR expression has a key role in propensity to consume caffeine . An interaction between CYP1A2 and AHR could be biologically plausible; however , we did not find any evidence supporting statistical interaction between the top two loci ( data not shown ) . Human and animal candidate gene studies for caffeine intake and related traits have focused on various other genes linked to caffeine's metabolism and targets of action . In our candidate gene analyses , we observed significant gene-based associations between CYP2C9 and ADORA2A and caffeine intake in addition to CYP1A2 and AHR . CYP2C9 catalyzes the N7-demethylation and C8-hydroxylation of caffeine to theophylline and 1 , 3 , 7-trimethyluric acid ( a minor metabolite ) , respectively; but its role relative to CYP1A2 is generally small[10] . In amounts typically consumed from dietary sources , caffeine antagonizes the actions of adenosine at the adenosine A2A receptor ( ADORA2A ) [2] , which plays an important role in the stimulating and reinforcing properties of caffeine [14] , [15] . Polymorphisms of ADORA2A have been previously implicated in caffeine-induced anxiety as well as habitual caffeine intake[16] , [17] . All studies contributing to our GWAS of caffeine intake were US-based . Consistent with the adult caffeine consumption pattern of this country , coffee contributed to well over 80% of caffeine intake . Previous studies suggest that some of the heritability underling specific caffeine sources ( i . e . coffee and tea ) may be distinct in relation to total caffeine intake [18] . To evaluate the robustness of findings , we conducted an additional GWAS analysis using caffeinated coffee intake as the outcome variable yielding the same strong signals ( rs4410790: 1 . 4×10−29 , rs2470893: 3 . 6×10−19 ) . Imprecision in phenotypic assessment and differences across studies could have limited the scope of our discovery . Although dietary intake obtained by FFQ is subject to misclassification , validation studies in subsamples of the included studies indicated that the consumption of caffeine-containing beverages is assessed with good accuracy [19] , [20] , [21] . The cubic root transformation we applied to reported caffeine intakes , however , limits interpretation of the effect estimates . The crude weighted mean difference in caffeine intake between homozygote genotypes was 44 mg/d for rs4410790 and 38 mg/d for rs2470893 ( Table S3 and S4 ) . The two SNPs together , however , explained between 0 . 06 and 0 . 72% of the total variation in caffeine intake across studies suggesting additional variants remain to be discovered [22] . Finally , our GWAS assumed an additive genetic model and based on study-level results ( Figure 1 and Figure 2 ) potential non-linear effects will require confirmation in future studies . Caffeine intake has been associated with pleotropic physiologic effects in relation to both detrimental and beneficial health outcomes [23] . Our current study provides insights into the primary pathways underlying caffeine intake . Knowledge of the genetic determinants of caffeine intake may provide insight into underlying mechanisms and may provide ways to study the potential health effects of caffeine more comprehensively by using genetic determinants as instrumental variables for caffeine intake or by taking into consideration caffeine-gene interactions . With the exception of nicotine dependency and the associated nicotinic receptor , genes that influence traits associated with dependency have been difficult to identify . The association of caffeine consumption with genes involved in metabolism or its regulation ( CYP1A2 and AhR , respectively ) illustrates that it is feasible to use GWAS to identify genetic determinants of other behavioral traits that are assessed with lower accuracy . We also recognize that the identified variants could influence regulation of their genomic elements distant from the known , high profile , neighboring candidate genes . In conclusion , we identified two loci related to caffeine consumption that will be worthy of further investigation with regard to both beneficial and toxic effects of caffeine as well as the extensive group of carcinogens , drugs , and xenobiotics also metabolized through action of the regulation of the gene products of CYP1A2 and AHR .
This study was conducted according to the principles expressed in the Declaration of Helsinki . All participants in the contributing studies gave written informed consent including consent for genetic analyses . Local institutional review boards approved study protocols . We conducted a meta-analysis of 47 , 341 individuals of European descent , sourced from Atherosclerosis Risk in Communities ( ARIC , N = 8 , 976 ) , the Prostate , Lung , Colorectal , and Ovarian Cancer Screening Trial ( PLCO , N = 4 , 942 ) , the Nurses' Health Study ( NHS , N = 6 , 774 ) , the Health Professionals Follow-Up Study ( HPFS , N = 4 , 023 ) , and the Women's Genome Health Study ( WGHS , N = 22 , 658 ) to identify novel loci associated with habitual caffeine consumption . Study population descriptions and genotyping quality control for data generated with either the Affymetrix 6 . 0 or the Illumina Infinium arrays ( HumanHap300 , 550 or 610 arrays ) are provided in Text S1 and Table S5 and S6 . In the NHS , every 2 to 4 years of follow-up diet was assessed using a validated semi-quantitative food frequency questionnaire ( FFQ ) [24] . For the present analysis , we included the participants' mean caffeine intakes of the 1984 ( first year in which caffeinated and decaffeinated coffee were differentiated ) and 1986 FFQs . The following caffeine-containing foods and beverages were included in the FFQ: coffee with caffeine , tea , cola and other carbonated beverages with caffeine , and chocolate . For each item , participants were asked how often , on average , they had consumed a specified amount of each beverage or food over the past year . The participants could choose from nine frequency categories ( never , 1–3 per month , 1 per week , 2–4 per week , 5–6 per week , 1 per day , 2–3 per day , 4–5 per day and 6 or more per day ) . Intakes of nutrients and caffeine were calculated using US Department of Agriculture food composition sources . In these calculations , we assumed that the content of caffeine was 137 mg per cup of coffee , 47 mg per cup of tea , 46 mg per can or bottle of cola or other caffeinated carbonated beverage , and 7 mg per 1 oz serving of chocolate candy . We assessed the total intake of caffeine by summing the caffeine content for the specified amount of each food multiplied by a weight proportional to the frequency of its use . In a validation among a subsample of this cohort , we obtained high correlations between intake of caffeinated coffee and other caffeinated beverages from the FFQ and four 1-week diet records ( coffee , r = 0 . 78; tea , r = 0 . 93; and caffeinated sodas , r = 0 . 85 ) [21] . In the WGHS , caffeine intake was assessed at baseline ( 1991 ) using the same FFQ and caffeine algorithm as the NHS [25] . HPFS participants have been followed with repeated FFQs every 4 years . Caffeine-intake was assessed by the same methods as described above for the NHS cohort . In a validation study in a subsample of participants , we obtained high correlations between consumption of coffee and other caffeinated beverages estimated from the FFQ and consumption estimated from repeated 1-wk diet records ( coffee: r = 0 . 83; tea: r = 0 . 62; low-calorie caffeinated sodas: r = 0 . 67; and regular caffeinated sodas: r = 0 . 56 ) [21] . For the present analysis , we included the participants mean caffeine intakes of the 1986 ( baseline ) and 1990 FFQs . In the ARIC study , caffeine consumption was quantified at the baseline ( 1987–1989 ) examination from an interview-administered 66-item semi-quantitative FFQ[19] , [20] . The Harvard Nutrition Database was used to assign caffeine ( and nutrient ) content to each of the food and beverage line items . Line items quantifying consumption of caffeine-containing beverages included sodas ( regular and diet ) , coffee , and tea . The frequency of consumption of each of these items was multiplied by their caffeine content and summed across all beverages to obtain a total caffeine intake value . Caffeine intake in the PLCO trial was assessed at the randomization phase ( between 1992–2001 ) using responses from a FFQ developed at the National Cancer Institute called the Diet History Questionnaire ( DHQ ) . The DHQ was previously validated against four 24 hour dietary recalls [26] and asks about consumption frequency of 124 food items over the past 12 months , including the primary sources of caffeine: coffee , tea , and soft drinks . For soft drinks , participants selected among 10 possible frequency response categories from “never” to “6+ times per day , ” with three possible portion size response categories: <12 ounces or <1 can or bottle; 12–16 ounces or 1 can or bottle; or >16 ounces or >1 can or bottle . Frequency and portion size for coffee and tea were queried together as cups per unit time ranging from “none” to “6 or more cups per day . ” For all three of the above beverages , participants were asked the proportion of the time each were consumed in decaffeinated form ( almost never or never , about ¼ of the time , about ½ the time , about ¾ of the time , almost always or always ) . From these responses daily consumption of caffeine was computed taking into account the caffeine content , portion size , and frequency of intake . Caffeine estimates were derived from two 24-hour dietary recalls administered in the 1994-96 Continuing Survey of Food Intake by Individuals ( CSFII ) [27] , a nationally representative survey conducted during the period when the DHQ was being administered . Individual foods/beverages reported on the recalls were placed in food groups consistent with items on the DHQ and weighted mean nutrient values based on survey data were derived for adults stratified by sex using methods previously described [28] . Each study used either MACH [29] ( ARIC , NHS , HPFS , WGHS ) or IMPUTE [30] ( PLCO ) to impute up to ∼2 . 5 million autosomal SNPs with NCBI build 36 of Phase II HapMap CEU data ( release 22 ) as the reference panel . Genotypes were imputed for SNPs not present in the genome-wide arrays or for those where genotyping had failed to meet the quality control criteria . Imputation results are summarized as an “allele dosage” ( a fractional value between 0 and 2 ) , defined as the expected number of copies of the minor allele at that SNP . The algorithm used for the calculation of caffeine intake was study-specific to allow for differences in questionnaires and consumption habits in different study populations . Raw caffeine-intake measures were skewed across studies and after exploring a variety of transformation options , we found that a cubic-root transformation was very close to the most optimal transformation identified by the Box-Cox procedure and was used to ensure normality of the residuals . Our final models were also adjusted for age ( continuous ) , sex , case-control status ( if applicable ) , study-site ( if applicable ) , smoking status ( never , former , and current: 2 categories ) , and study specific eigenvectors ( see Table S5 for study-specific models ) . Adjustment for smoking status was appropriate given the strong correlation between smoking and caffeine intake that might impede our ability to uncover caffeine-specific loci . Each study collected information on smoking status at the time FFQ were administered . A flexible modeling approach was used to accommodate the different methods by which smoking was collected across studies , but all included never , former and two categories of current smokers . Further adjustments for body-mass-index did not change results appreciably . Each study performed genome-wide association testing for normalized caffeine-intake across ∼2 . 5 million SNPs , based on linear regression under an additive genetic model . Analyses were adjusted for additional covariates as described above and further detailed in Table S5 . Imputed data ( expressed as allele dosage ) were examined using ProbABEL[31] or R ( scripts developed in-house ) . The genomic inflation factor λ for each study as well as the meta-analysis was estimated from the median χ2 statistic . Meta-analysis was conducted using a fixed effects model and inverse-variance weighting as implemented in METAL ( see URLs in Text S1 ) . The software also calculates the genomic control parameter and adjusts each study's standard errors . Fixed effects analyses are regarded as the most efficient method for discovery in the GWAS setting [32] . Heterogeneity across studies was investigated using the I2 statistic[33] . We applied stringent quality filters to imputed SNPs prior to meta-analysis; removing those with <0 . 02 MAF and/or with low imputation quality scores . The latter was defined as Rsq≤0 . 80 for SNPs imputed with MACH and proper_info≤0 . 7 for SNPs imputed with IMPUTE . X and Y chromosome , pseudosomal and mitochondrial SNPs were not included for the present analysis . We retained only SNP-phenotype associations that were based on results from at least 2 of the 10 participating studies and if greater than 50% of the samples contributing to the results were genotyped . Additional checks for experimental biases were implemented for notable associations including manual inspection of SNP ( if imputed , an assayed SNP in high LD ) cluster plots , and evaluation of HWE , and comparison of study MAFs to the HapMap CEPH panel . We considered P-values <5×10−8 to indicate genome-wide significance [34] . We examined 515 SNPs in 23 genes ( ±50 kb ) either previously studied or members of the key biological pathway: ‘Caffeine metabolism’ ( KEGG [35] , supplemented with candidates from[10] , [36] ) for association with caffeine consumption in our GWA meta-analysis sample . SNPs mapping to TAS2R10 , 43 and 46 , implicated in the oral detection of caffeine , did not pass our stringent QC criteria and thus were not included . Gene-based analyses were performed using VEGAS [37] . The software applies a test that incorporates information from a set of markers within a gene ( or region ) and accounts for LD between markers by using simulations from the multivariate normal distribution . The number of simulations per gene is determined adaptively . In the first stage , 1000 simulations are performed . If the resulting empirical P value is less than 0 . 1 , 10000 simulations are performed . If the empirical P value from 10000 simulations is less than 0 . 0001 , the program will perform 1000000 simulations . At each stage , the simulations are mutually exclusive . For computational reasons , if the empirical P value is 0 , then no more simulations will be performed . An empirical P value of 0 from 1000000 simulations can be interpreted as P<10 E-6 , which exceeds a Bonferroni-corrected threshold of P<2 . 8E-6 [∼0 . 05/17 , 787 ( number of autosomal genes ) ] . | Caffeine is the most widely consumed psychoactive substance in the world . Although demographic and social factors have been linked to habitual caffeine consumption , twin studies report a large heritable component . Through a comprehensive search of the human genome involving over 40 , 000 participants , we discovered two loci associated with habitual caffeine consumption: the first near AHR and the second between CYP1A1 and CYP1A2 . Both the AHR and CYP1A2 genes are biologically plausible candidates , as CYP1A2 metabolizes caffeine and AHR regulates CYP1A2 . Caffeine intake has been associated with manifold physiologic effects and both detrimental and beneficial health outcomes . Knowledge of the genetic determinants of caffeine intake may provide insight into underlying mechanisms and may provide ways to study the potential health effects of caffeine more comprehensively . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] | [
"social",
"and",
"behavioral",
"sciences"
] | 2011 | Genome-Wide Meta-Analysis Identifies Regions on 7p21 (AHR) and 15q24 (CYP1A2) As Determinants of Habitual Caffeine Consumption |
Here we put forward a mathematical model describing the response of low-grade ( WHO grade II ) oligodendrogliomas ( LGO ) to temozolomide ( TMZ ) . The model describes the longitudinal volumetric dynamics of tumor response to TMZ of a cohort of 11 LGO patients treated with TMZ . After finding patient-specific parameters , different therapeutic strategies were tried computationally on the ‘in-silico twins’ of those patients . Chemotherapy schedules with larger-than-standard rest periods between consecutive cycles had either the same or better long-term efficacy than the standard 28-day cycles . The results were confirmed in a large trial of 2000 virtual patients . These long-cycle schemes would also have reduced toxicity and defer the appearance of resistances . On the basis of those results , a combination scheme consisting of five induction TMZ cycles given monthly plus 12 maintenance cycles given every three months was found to provide substantial survival benefits for the in-silico twins of the 11 LGO patients ( median 5 . 69 years , range: 0 . 67 to 68 . 45 years ) and in a large virtual trial including 2000 patients . We used 220 sets of experiments in-silico to show that a clinical trial incorporating 100 patients per arm ( standard intensive treatment versus 5 + 12 scheme ) could demonstrate the superiority of the novel scheme after a follow-up period of 10 years . Thus , the proposed treatment plan could be the basis for a standardized TMZ treatment for LGO patients with survival benefits .
Oligodendrogliomas are low-incidence glial tumors , affecting mostly young adults . They are slowly growing , infiltrative tumors with isocitrate dehydrogenase 1 or 2 mutations and codeletion of chromosomal arms 1p and 19q . Grade II oligodendrogliomas ( LGOs ) are well differentiated tumors with a low mitotic index [1] . In spite of the long median patient survival , they are incurable currently [2] . Many oligodendroglioma patients present few neurological symptoms for extended periods of time . The decision on the specific combination of therapies to be used on each patient is based on the qualitative consideration of different variables including age , tumor grade , performance status and tumor location [3] . Radiation therapy ( RT ) is beneficial for patients in terms of survival , but its timing has been the subject of debate [4] . Regarding chemotherapy , temozolomide ( TMZ ) , an oral alkylating agent , has a favourable toxicity profile [5] and can contribute to reduction in seizure frequency in low-grade glioma patients [6] . Phase II trials have demonstrated its effectivity against low grade gliomas [7–9] . Also , neoadjuvant chemotherapy given to surgically unresectable tumors has allowed subsequent gross total resection in some cases [10] , which is of relevance when the tumour is highly infiltrative or located in eloquent areas . Thus , prolonged TMZ treatment is a relevant option either as up-front or as adjuvant treatment . Clinical trials have shown a similar efficacy of TMZ vs RT for 1p/19q-codeleted tumors [11 , 12] . Also , RT is associated with late neurocognitive toxicity . Thus chemotherapy is frequently used as first-line treatment for oligodendroglioma patients . In that context , relevant questions arise such choice of the chemotherapy regimen and the optimal number of cycles to be prescribed . Mathematical models have potential to help in finding optimized treatment schedules/combinations improving survival and/or reducing toxicity [13 , 14] . Once the base mathematical model is set , patient-specific parameters can be obtained from data . That provides an ‘in-silico twin’ [15] allowing computational studies that could be beneficial for real patients .
The study was approved by Kantonale Ethikkommission Bern ( Bern , Switzerland ) , with approval number: 07 . 09 . 72 . 82 patients diagnosed with low-grade gliomas ( biopsy/surgery confirmed astrocytoma , oligoastrocytoma or oligodendroglioma according to the WHO 2007 classification ) and followed at the Bern University Hospital between 1990 and 2013 were initially included in the study . Of that patient population , we selected grade II 1p/19q-codeleted tumors ( thus LGOs according to the 2016 WHO classification ) 36 patients . 17 of those patients did not receive TMZ and 2 received TMZ in combination with other therapy . Finally two of them received TMZ only after becoming anaplastic tumors . Of the remaining 16 patients , three did not respond to TMZ and two responded initially but progressed during treatment to anaplastic forms . Thus 11 oligodendroglioma patients treated with at least three cycles of TMZ , responded to the therapy , had neither previous RT treatment nor other treatment given in the period of study and did not display any signs of malignant transformation . Radiological glioma growth was quantified by manual measurements of tumour diameters on successive MRIs ( T2/FLAIR sequences ) . Since some of the older patients were available only as jpeg images we computed the volume using the ellipsoidal approximation . The three largest tumour diameters ( D1 , D2 , D3 ) along the axial , coronal and sagittal planes were measured and tumour volumes estimated using the equation V = ( D1 ⋅ D2 ⋅ D3 ) /2 , following the standard practice [16] . To estimate the error of the methodology we took a different set of glioma patients from another study [17] and compared their volumes computed accurately using a semi-automatic segmentation approach with those computed using the ellipsoidal approximation . Mean differences were 18% , that was the reference level used for the error in the volume computations . In this paper we considered LGOs in a simplified way as composed of two tumor cell compartments . The first one was the tumor cell population P ( t ) , assumed to grow logistically . The second one was lethally damaged tumor cells because of the action of the therapy D ( t ) . Temozolomide effect on proliferative cells is a complex one , leading to death through different ‘programmes’ [18–20] . We put together the different processes into two groups , each described by a term in our equations . The first one was early death accounting for necrosis , autophagy and drug-induced apoptosis with rate α1 . The second one was delayed death through mitotic catastrophe with rate α2 . The drug concentration in tissue was described by the function C ( t ) with an elimination rate constant , λ . Fig 1 shows a schematic description of the model . The equations were: d P d t = ρ P ( 1 - P + D K ) - α 1 P C - α 2 P C , ( 1a ) d D d t = - ρ κ D ( 1 - P + D K ) + α 1 P C , ( 1b ) d C d t = - λ C , ( 1c ) Chemotherapy was described by a sequence of doses d given at times t1 < t2 < … < tN . The initial time corresponding to the first volumetric observation was denoted as t0 . Initial conditions for Eq ( 1 ) were taken to be P0 = P ( t0 ) , D ( t0 ) = C ( t0 ) = 0 . Let us define f ( t j - ) = lim t → t j - f ( t ) and C is the fraction of the dose d reaching brain tumor tissue , assumed to to be estimated later . Drug administration was described as impulses for the times tj so that P ( t j ) = P ( t j - ) , D ( t j ) = D ( t j - ) , C ( t j ) = C ( t j - ) + C . Since cell numbers are expected to be proportional to the volumes occupied by each tumor subpopulation we worked with the later quantities , that are directly measurable . Thus , P ( t ) and D ( t ) through this paper were measured in cm3 . We chose the parameter κ , corresponding to the averaged number of cell divisions before death by mitotic catastrophe to be equal to 1 . In mitotic catastrophe , damaged cells do not die until they try to perform mitosis again . Although a few cells may be able to complete more than one cycle , most of them will die in the first one [21] . This is a reasonable assumption allowing us to get rid of this parameter . The carrying capacity parameter K is the one with a less defined value but could be expected to be in a range between 300 and 550 cm3 . The later number is in line with the maximal volumes observed in low-grade glioma patients [22] . However , many patients die when the tumor volume is smaller [15] . The most typical chemotherapy schedule consists of cycles of 28 days with five TMZ oral doses on days 1 to 5 and then a rest period of 23 days . Typical dose per day is d = 150 mg per m2 of patient body surface . To calculate the rate of drug decay λ we followed the same methodology as in Ref . [26] , using values of TMZ half-life clearance time t1/2 for doses of 150 mg/m2 . From the definition of t1/2 and since Eq ( 1c ) has exponentially decaying solutions we obtain that 1/2 = exp ( −λt1/2 ) . To account also for the drug loss during transport to the brain we computed the value of the dose getting to the tumour as C = β ⋅ d ⋅ b , where β is the fraction of TMZ getting to 1 cm3 of brain interstitial fluid ( from a unit dose ) and b is the patient’s body surface . Then C0 can be interpreted as an effective dose per fraction . The parameter β can be calculated using the value of maximal TMZ concentration Cmax for a dose of 150 mg/m2 taken from the literature [23] . Since time to reach peak drug concentration in brain is smaller than two hours and thus negligible in comparison with the other time scales in the model , we chose to set the initial drug concentration C0 to the value Cmax = 0 . 6 μg/cm3 as in Ref . [26] . The parameters α1 , α2 and ρ are expected to depend strongly on the tumor growth rate and sensitity to the therapy and will be considered to be adjustable parameters . These parameters , together with the initial population value P ( 0 ) were fit for each patient’s longitudinal volumetric data using the library fmincon in the scientific software package Matlab ( R2017b , The MathWorks , Inc . , Natick , MA , USA ) . Table 1 summarizes the main characteristics and parameter values found for patients included in the study . The fits obtained with those parameters were accurate , with a low relative error as computed as the difference between the tumor volume data and those obtained by the best fit . Mathematically e ¯ = ( 100 / N ) ∑ i = 1 N | V i p - V i f | / V i p ( expressed in percent ) , where Vip was tumour volume data and Vif was the tumour volume calculated with the parameters obtained by the least square fitting . The average of the relative errors for all the simulations was 9 . 1% and the median 8 . 3% , that were within the volume measurement uncertainty . Numerical simulations of Eq ( 1 ) were performed using the Matlab library ode45 . Results for the parameters are listed in Table 1 . To test the effect of different drug schedules for each virtual patient obtained from our data we performed series of numerical simulations . A first set of simulations consisted of a schedule were five daily doses of the drug were given on days 1-5 of the cycle and then the standard waiting period of 23 days was increased with variable times of up to 6 months with intervals of 15 days , i . e . 12 different spacings were studied for each virtual patient . These strategies will be denoted to as ‘long-cycle’ ones in what follows . Other sets of simulations consisted on the redistribution of the five doses during the 28 days of the cycle in two different ways , to be denoted hereafter as ‘distributed dose’ regimens . A first distributed regime consisted in 5 doses given following a 1-day on , 1-day off scheme during the first 10 days of the cycle . A second alternative was distributing doses evenly within the cycle duration , i . e . giving a single dose every 4 days . To study the effect of the different treatment schedules on patient survival we designed virtual trials . A number of virtual patients was generated by a random choice of the parameters . Uniform distributions were taken for the parameters in the most representative region of the parameter space obtained from Table 1: ρ ∈ [0 . 5 × 10−3 , 2 . 5 × 10−3] day- 1 , α1 ∈ [0 . 01 , 1 . 0] cm3/μg day , α2 ∈ [0 . 1 , 0 . 75] cm3/μg day , P ( 0 ) ∈ [20 , 200] , K ∈ [300 , 550] cm3 . Parameters were randomly sampled from the distributions in order to perform the simulations . 2000 patients were included in each virtual trial ( 1000 patients on each of the two arms considered ) . Virtual trials were run using Matlab 2017b parallel computing toolbox using a parallel algorithm on a 64 GB memory 2 . 7 GHz 12-core Mac pro workstation under OS X 10 . 14 . For the survival studies patients were assumed to die when tumors reached a volume of 280 cm3 and those alive after 25 years were considered as censored events . That volume would correspond to a sphere of 8 cm in diameter . Previous theoretical studies of glioblastoma growth have found the fatal tumor burden to be around 7 cm in diameter [15] . Since low-grade gliomas are less aggressive and harmful for the brain we chose the lethal size to be somewhat larger . In real life that number would obviously depend on tumor location , aggressiveness , patient overall status , etc .
Typical low-grade glioma longitudinal growth and response to therapy consists of four stages ( see Fig 2 ) . First , without treatment tumor grows slowly but steadily [24] . Next , there is an early ‘fast’ tumor volume reduction associated to the start of treatment with TMZ . Finally , after treatment cesation , there is a long-term response . For the patient shown in Fig 2 , the tumor volume reduction lasted for 14 months after the end of the treatment course . Finally , the tumor regrew leading to a clinical relapse . All of those stages were described by the mathematical model . Each stage was associated to one of the biological phenomena reflected as terms in the model equations . We studied the ability of our mathematical model to describe the tumor responses to TMZ . To do so , we fitted the parameters in Eq ( 1 ) using the longitudinal volumetric data for each patient in our cohort . Fig 3 shows results for selected patients . The model described the longitudinal tumor volumetric data in all cases . This validates the choice of biological mechanisms used in constructing the model . Results shown in Fig 3 were obtained for a fixed ( i . e . non fitted ) value of the carrying capacity K = 523 . 6 cm3 . This parameter provides an estimate of the tumor size for which geometrical and other constraints have a substantial influence on the tumor growth rate . Specifically , this is the volume of a sphere of radius 5 cm , a number that is about 2/5 of the whole brain volume [25] . A tumor of this size would be incompatible with the patient’s life and very difficult to obtain given the brain physical barriers . This is the K value that will be taken for most of the paper simulations . However , similar results were obtained for a broad range of values of K . As an example , Fig 3 ( g ) –3 ( i ) shows results for selected patients using a smaller K = 261 . 8 cm3 . Indeed , the value of K cannot be uniquely determined from the data . Different values of K lead to slightly different fitting parameter choices and similar shapes of the fitting curves . Thus we chose to fix it through the paper . However the results to be presented do not depend on the specific choice of K . The model was then used as a discovery platform to test alternative treatment regimens in-silico for the patients included in the study . As a first test , we enlarged the time interval between cycles . Five daily doses of the drug were given on days 1-5 of the cycle and then the standard waiting period of 23 days was increased as described in methods . In general , the long-term tumor evolution was similar for all the schedules when the cycle’s length was in the range 1-4 months . Thus , from the volumetric point of view , all schedules led to similar asymptotic dynamics for the virtual patients . Results for selected patients are shown in Fig 4 . Long-cycle treatment regimens resulted in smaller tumor volume reduction due to the less intensive nature of the schemes . For small tumors , the nonlinear terms in Eq ( 1 ) played a marginal role as far as the values of ( P + D ) /K were small . However , in the case of large tumors , whose P + D size was comparable to K , the nonlinear terms had a substantial contribution to the dynamics ruled by Eq ( 1 ) . In the latter case , differences between the schemes were observed favoring long-cycle schemes ( see e . g . Fig 4 ( b ) ) . As a second series of tests , we explored alternative treatment regimens based on the 28-day cycle , the distributed dose regimens as described in ‘Methods’ . All distributed treatment regimens led to tumor volumetric evolutions equivalent to the ones of the standard treatment ( e . g . , those depicted in Fig 3 ) . Several virtual trials were conducted as described in the ‘Methods’ section . Benefits in median survival were found for the long-cycle strategies that were dependent on the parameter K . Long-cycle treatment schemes were never inferior in terms of survival to the standard ones . Indeed , the differences found between survival curves for long-cycle schedules versus the standard ones were never statistically significant ( p < 0 . 05 ) according to the log-rank test . Patients in our retrospective dataset were treated with a variable number of TMZ cycles ( mean 12 , range 4-20 , see Table 1 ) . Treatment was effective for all patients included . However , since there is no standardized protocol for chemotherapy in LGO patients , the decision to stop treatment was taken depending on toxicity , physician and patient preferences , etc . Next we explored the potential effectiveness of standardizing treatment for the virtual patients obtained from our patient’s data in-silico . To do so we tested our scheme consisting of an induction part of five cycles given monthly to substantially reduce the tumor burden followed by a consolidation of 12 cycles given every three months . This treatment scheme was based on the idea that TMZ cycles given every three months should be well tolerated and allow for this long schedule . Moreover , having a first induction part would result in an initial larger tumor volume reduction than for the long-cycle schemes alone . Results are summarized in Fig 5 . Survival improvements , many of them substantial , were obtained for the virtual counterparts of the patients included in the study ( Median 5 . 69 years , range: 0 . 67 to 68 . 45 years , see Fig 5 ( a ) ) . Virtual patients for which the number of cycles was larger ( patients 3 , 6 , 7 , 8 and 10 ) than those received by the real one ( see Table 1 ) had larger survival benefits . Also for most patients there was a substantial volumetric reduction in relation to the one achieved for the real patient under the number of cycles given by Table 1 ( see Fig 5 ( b ) ) . Although those results are interesting , they hold for a limited number of patients . Thus , to complement the previous results we developed a virtual trial with 2000 virtual patients included in two arms as described in ‘Methods’ . Results are summarized in Fig 6 . Differences between the curves were statistically significant ( p = 1 . 65 × 10−14 , HR = 0 . 679 ( 0 . 614−0 . 75 ) ) , with a difference in median survival of 3 . 8 years between both treatment arms . The main interest of our results is the possibility of the proposed scheme to become the basis of a clinical trial that could be beneficial for LGO patients . However , real clinical trials are limited in size ( number of patients included ) and follow-up time . To test the feasibility of a confirmatory clinical trial we performed a series of numerical experiments , whose results are summarized in Fig 7 . For an initial set of 20 patients per arm we compared the standard intensive scheme versus the 5+12 one within a followup period of 10 years and compared the survival of both populations . The experiment was repeated 20 times by resampling randomly the parameter distribution as described in ‘Methods’ and in only 5 of the trials ( 25% ) the differences between treatment arms achieved statistical significance p < 0 . 05 . Then the patient population was increased in intervals of 10 patients until a final population of 110 per treatment arm . For each patient population , 20 trials were developed in-silico . As the patient number increased the statistical significance of the trials was better . For the 20 trials performed with 100 patients per arm and 110 patients per arm all p-values were below the statistical significance limit p = 0 . 05 ( Fig 7 ) . Thus a trial with more than 100 patients per arm with 10 years followup should be able to show the superiority of the 5+12 scheme over the standard one with a high confidence ( >97 . 5% ) .
We developed a mathematical model of LGOs response to TMZ describing the longitudinal tumor volumetric dynamics . Once fitted for each patient , the model provided a set of parameters describing the behavior of each of the real patients . When subjected to long-cycle treatment regimens the virtual patients showed similar or better performance in terms of survival . In-silico clinical trials confirmed the results for broader parameter regimens . This long-cycle TMZ schedules could prove beneficial for LGO patients in terms of toxicity . We studied ‘in-silico’ a treatment combining an induction phase of 5 consecutive cycles plus a maintenance phase ( 12 cycles given in three-months intervals ) . The improved drug-exposure of this scheme led to substantial survival improvements and a good tumor control in-silico . We hope this computational study could provide a theoretical ground for the development of clinical studies and the definition of standardized TMZ treatment protocols for low-grade oligodendroglioma patients with improved survival . | We developed a mathematical model describing the longitudinal volumetric growth data of grade II oligodendroglioma patients and their response to temozolomide . The model was used to explore alternative therapeutic protocols for the in-silico twins of the patients and in virtual clinical trials . The simulations show that enlarging the time interval between chemotherapy cycles would maintain the therapeutic efficacy , while limiting toxicity and deferring the development of resistance . This may allow for improved drug-exposure by administering a larger number of cycles for longer treatment periods . A scheme based on this idea consisting of an induction phase ( 5 consecutive cycles , 1 per month ) and a maintenance phase ( 12 cycles given in three-months intervals ) led to substantial survival benefits in-silico . The computational results suggest that a clinical trial enrolling 100 patients per arm ( standard intensive therapy versus 5+12 novel scheme ) could prove the effectiveness of the proposed approach after a follow-up period of 10 years . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"cancer",
"treatment",
"clinical",
"oncology",
"cell",
"processes",
"cancers",
"and",
"neoplasms",
"mathematical",
"models",
"toxicology",
"oncology",
"neurological",... | 2019 | Computational design of improved standardized chemotherapy protocols for grade II oligodendrogliomas |
Eukaryotic gene expression involves tight coordination between transcription and pre–mRNA splicing; however , factors responsible for this coordination remain incompletely defined . Here , we explored the genetic , functional , and biochemical interactions of a likely coordinator , Npl3 , an SR-like protein in Saccharomyces cerevisiae that we recently showed is required for efficient co-transcriptional recruitment of the splicing machinery . We surveyed the NPL3 genetic interaction space and observed a significant enrichment for genes involved in histone modification and chromatin remodeling . Specifically , we found that Npl3 genetically interacts with both Bre1 , which mono-ubiquitinates histone H2B as part of the RAD6 Complex , and Ubp8 , the de-ubiquitinase of the SAGA Complex . In support of these genetic data , we show that Bre1 physically interacts with Npl3 in an RNA–independent manner . Furthermore , using a genome-wide splicing microarray , we found that the known splicing defect of a strain lacking Npl3 is exacerbated by deletion of BRE1 or UBP8 , a phenomenon phenocopied by a point mutation in H2B that abrogates ubiquitination . Intriguingly , even in the presence of wild-type NPL3 , deletion of BRE1 exhibits a mild splicing defect and elicits a growth defect in combination with deletions of early and late splicing factors . Taken together , our data reveal a connection between Npl3 and an extensive array of chromatin factors and describe an unanticipated functional link between histone H2B ubiquitination and pre–mRNA splicing .
Pre-mRNA splicing is a critical step in gene expression in which non-coding introns are removed from pre-mRNA and protein-coding exons are ligated together . This process is performed by the spliceosome , a dynamic ribonucleoprotein particle that , in yeast , consists of 5 snRNAs and over 80 proteins that cooperate to recognize and splice target mRNAs [1] . Recent evidence reveals that mRNA splicing in vivo is largely co-transcriptional , and occurs while elongating RNA polymerase II ( PolII ) is still associated with chromatin [2]–[4] . The basic unit of chromatin is 146 base pairs of DNA wound around a histone octamer to form a nucleosome , arrays of which can be further compacted to form higher-order chromatin structure . A plethora of chromatin remodeling and histone modifying machines are now known to be integral parts of the gene expression process [5] . While much has been learned about the molecular mechanisms of pre-mRNA splicing from in vitro systems [6] , a full understanding of the regulation of spliceosome assembly and catalysis will require an appreciation of the complex landscape of the chromatinized template , along which splicing occurs . To approach this question , we built upon our recent observation that the SR-like protein Npl3 promotes efficient splicing of a large subset of genes via co-transcriptional recruitment of U1 and U2 snRNPs [7] . SR and hnRNP proteins in metazoa are best understood for their role in alternative and constitutive splicing , although they have also been implicated in additional steps in gene expression , including mRNA export , translation , and even transcription itself [8]–[10] . Despite the fact that there are few examples of alternative splicing in S . cerevisiae , this yeast contains three genes with a canonical SR protein domain structure: one or more RNA recognition motifs and a domain enriched in arginine-serine dipeptides [9] , [11] . We recently demonstrated that deletion of NPL3 specifically , but not the others , impacts splicing; interestingly , the affected genes are almost exclusively those encoding ribosomal proteins , and make up the largest class of intron-containing genes in budding yeast [7] . Npl3 appears to be appropriately poised to coordinate events in gene expression: it is recruited to chromatin early during transcription [12] stimulates transcription elongation [13]–[16] co-purifies with elongating PolII [12] , [14] via its interaction with the C-terminal domain [14] , and remains associated with mRNA after processing is completed [17] , [18] . Here , in order to understand how Npl3 might choreograph gene expression events in S . cerevisiae , we systematically analyzed genetic interactions of a strain lacking Npl3 . We uncovered genetic interactions between the npl3Δ allele and genes involved in transcription and chromatin modification , including factors involved in histone H2B ubiquitination: the E3 ubiquitin ligase , Bre1 [19] , [20] , and corresponding ubiquitin protease , Ubp8 [21]–[23] . In addition , we show that Npl3 physically interacts with Bre1 . Splicing-sensitive microarray experiments reveal that disabling the H2B ubiquitination pathway by deletion of BRE1 or UBP8 , or point mutation of H2B , exacerbates the known splicing defect of an npl3Δ strain . Furthermore , we observed an Npl3-independent connection between Bre1 and splicing , as deletion of BRE1 impairs the splicing of a subset of pre-mRNAs and , in combination with deletions of individual splicing factors , causes severe synthetic growth defects . Thus , our data functionally link H2B ubiquitination by Bre1 to pre-mRNA splicing and more broadly suggest that the coordination of transcription and splicing may be aided by crosstalk between Npl3 and chromatin metabolism .
The SR-like protein Npl3 has multiple roles in the regulation of gene expression , including in pre-mRNA splicing , 3′ end processing , and mRNA export . To further interrogate this multifunctional factor , we used synthetic genetic array ( SGA ) technology [24] , [25] to screen ∼4 , 800 non-essential yeast genes for those whose deletion conferred synthetic lethality ( SL ) or very synthetic sick ( SS ) growth phenotypes in an npl3Δ strain . Since an npl3Δ strain grows more slowly than wild-type at 30°C , and this defect is exacerbated at 37°C ( e . g . , see Figure 1C , top panels ) , we performed the screen at both temperatures to maximize coverage . The analysis revealed strong negative interactions between NPL3 and 83 ( 1 . 7% of total ) and 333 ( 6 . 9% of total ) genes after growth at 30°C and 37°C , respectively ( see Table S1 ) . To validate a subset of genetic interactions identified by this high-throughput approach , we generated the cognate double mutant strains using tetrad dissection . In order to refine our list of genetically interacting factors , we included additional subunits from complexes represented in the results of the screen . A list of the most stringent synthetic interaction partners ( identified in the 30°C SGA and directed genetics ) was integrated with those from a previously published quantitative RNA processing Epistatic Mini Array Profile ( E-MAP ) [26] to generate a more comprehensive set of NPL3 SS/SL genetic interactions ( Figure 1A and Materials and Methods ) . These negative genetic interactions were highly enriched for genes that function in RNA metabolism ( Table S2 ) , consistent with what was previously known about Npl3 function in mRNA processing [7] , [12] , [13] , [16]–[18] , [27]–[31] . In addition , there was an enrichment of genetic interactions with genes implicated in “chromosome organization” and “transcription , ” including components of the chromatin remodeling SWR Complex [32]–[34] , the transcriptional elongation PAF Complex [35]–[38] , and multiple histone modification complexes , including COMPASS [39] , SAGA [40] , and the SET3 Complex [41] ( Figure 1B , 1C and 1D and Table S3 ) . We note that of these , the SWR1 and SAGA Complexes have previously been implicated in pre-mRNA splicing [42] , [43] , highlighting the ability of the Npl3 screen to identify factors involved in chromatin-splicing crosstalk . The screens also showed that deletion of either RAD6 ( Figure 1E cf . closed triangles ) or BRE1 ( Figure 1E cf . open triangles ) led to synthetic sickness/lethality in an NPL3 deletion strain . These factors catalyze the mono-ubiquitination of lysine 123 on histone H2B; specifically , Bre1 is the E3 ubiquitin ligase and Rad6 is its corresponding E2 ubiquitin-conjugating enzyme [19] , [20] , [44] , [45] . We found that inactivating Bre1 ubiquitin ligase activity via a point mutation in its RING domain ( bre1H665A ) [19] exacerbated the growth defect of an npl3Δ strain to the same extent as a full deletion of BRE1 ( Figure 1E , cf . orange triangles ) , suggesting that the genetic interaction is connected to the ligase activity of Bre1 . Many nuclear enzymes act not only on histones but on other substrates as well , and , in fact , histone H2B is not the only ubiquitination target of Bre1 [46] . To ask whether the Npl3-Bre1 genetic interaction is due to the loss of H2B ubiquitination specifically , we tested whether a mutation of the target residue in H2B would phenocopy a deletion of BRE1 . Indeed , the htb1K123R point mutant also profoundly exacerbated the growth defect of npl3Δ ( Figure 1E cf . purple triangles ) . Taken together , these data provide strong evidence that H2B ubiquitination can account for the genetic interaction of the RAD6 Complex with NPL3 . The PAF Complex and COMPASS have previously been shown to function in the same histone modification pathway as the Bre1 [47]–[50] . The PAF Complex is required for H2B ubiquitination [49] , [50]; thus , the synthetic lethality we observed between NPL3 and components of the PAF Complex ( Figure 1C and 1D ) was consistent with the genetic interactions we observed with the Bre1 . H2B ubiquitination is , in turn , required for trimethylation of histone H3 lysine 4 ( H3K4 ) by COMPASS [39] , [51]–[53] and lysine 79 ( H3K79 ) by Dot1 [54]–[57] . However , we found no genetic interaction between NPL3 and point mutations of H3K4 or H3K79 ( data not shown ) , suggesting that loss of these chromatin marks is unlikely to underlie the synthetic sickness in the npl3Δbre1Δ double mutant . Given that maintaining H2B ubiquitination is critical in the absence of NPL3 , it follows that mutations in genes required for the removal of this chromatin mark might suppress the npl3Δ growth defect . To investigate this in an unbiased fashion , we made use of the fact that NPL3 deletion causes lethality when yeast are grown at 16°C ( e . g . , see Figure 2B , top panel ) ; this allowed us to screen for mutants that restore growth to an npl3Δ strain at 16°C . This screen identified 105 ( 2 . 1% of total ) and 699 ( 14 . 4% of total ) suppressors after 4 and 8 days of growth , respectively ( Table S4 ) , and a number of these suppressors have previously been implicated in transcription and chromatin modification ( Figure 2A ) . We then generated a number of the double mutants using tetrad dissection and validated the suppressive genetic interactions using serial dilution ( Figure 2B ) . In agreement with our expectation , the data from this screen showed that deletion of UBP8 , which encodes an H2B de-ubiquitinase [21]–[23] , restored viability to a strain lacking Npl3 ( Figure 2B cf . closed triangles ) . In further support of these observations , the SGA also identified SGF11 and SGF73 as genes whose deletion suppresses npl3Δ; these factors are part of a module of the SAGA Complex with Upb8 , and are also implicated in gene activation by H2B de-ubiquitination [21] , [58]–[61] . Taken together , this dataset shows that the npl3Δ strain is particularly sensitive to deletion of genes affecting the H2B ubiquitination pathway ( Figure 1 and Figure 2 ) and opens the possibility that H2B ubiquitination is important for an Npl3-dependent process . Interestingly , deletions of genes in other modules of SAGA required for either histone acetylation ( Ada2 and Gcn5 ) or for association of the SAGA complex with promoters ( i . e . , the TBP regulatory module , Spt3 and Spt8; reviewed in [62] ) exacerbated , rather than suppressed , the npl3Δ growth defect ( Figure 1C and Table S1 ) . The divergent genetic interactions confirm the functionally separable nature of the SAGA sub-modules [58] and highlight that a connection exists between Npl3 and H2B mono-ubiquitination that is functionally distinct from other chromatin marks . Given the robust genetic interactions we observed between NPL3 and genes involved in H2B ubiquitination , we performed co-immunoprecipitation assays of the corresponding proteins to test if they physically interact . We had previously shown that Npl3 co-immunoprecipitated components of the U1 snRNP [7] . Here , we immunoprecipitated endogenous Npl3 from whole-cell extract using a polyclonal antibody directed against Npl3 [63] and then probed the precipitate for endogenously tagged forms of Bre1 , Ubp8 , and Sgf11 as well as positive and negative controls ( a U1 protein , Luc7 , and Nup188 , respectively ) . Although there is precedent for some interaction specificity with the E3 Bre1 over the E2 Rad6 [46] , we also tested for an interaction with Rad6 . As shown in Figure 3 , only Bre1 and Luc7 but not Nup188 , Rad6 , Ubp8 or Sgf11 , co-immunoprecipitated with Npl3 . It is known that Npl3 is an RNA-binding protein , and its interaction with some components of the splicing machinery is RNA-dependent [7] . To test whether the observed interaction with Bre1 is mediated by RNA , we treated the extracts with RNaseA prior to the immunoprecipitation . We consistently found that a population of Bre1 interacts with Npl3 in an RNase-independent manner ( Figure 3 cf . lanes 3 and 4 , top panel ) . These data indicate that Npl3 can physically interact with Bre1 , consistent with previous data from high-throughput proteomic analyses [64] . The genetic data connecting NPL3 and the H2B ubiquitination machinery lend support for two possible models . One model predicts that Npl3 will affect H2B ubiquitination; we therefore measured the global percentage of ubiquitinated H2B but found the npl3Δ strain indistinguishable from wild-type ( Figure S1 ) . An alternative interpretation of the genetic data is that the H2B ubiquitination cycle is important for an Npl3-dependent process . We previously reported [7] that a strain lacking Npl3 accumulates a subset of pre-mRNAs , consisting primarily of the ribosomal protein genes ( RPGs ) , whose splicing efficiency might be expected to affect growth rate . Given that deletion of BRE1 exacerbates the npl3Δ growth defect , we tested whether deleting BRE1 exacerbates the npl3Δ splicing defect . We used our splicing-sensitive microarray platform [65] , which contains oligos that hybridize to the terminal exon , the intron , and the exon-exon junction of each intron-containing gene , in order to detect total mRNA , pre-mRNA , and mature mRNA , respectively ( Figure 4A ) . For each genotype , the heat map ( Figure 4B ) reports fold changes in signal intensity of these three RNA species for each intron-containing gene as compared to a wild-type strain . As expected , our experiments showed that a strain lacking Npl3 accumulated RPG pre-mRNAs ( Figure 4B npl3Δ , see yellow in Intron feature; RPGs highlighted in purple on right ) . Notably , the pre-mRNA accumulation in the npl3Δ strain was increased at many RPGs when BRE1 was also deleted ( Figure 4B cf . npl3Δ and npl3Δbre1Δ , Intron feature ) , suggesting that Bre1 is important for the splicing of many Npl3-dependent genes . We note that this effect is complex , and is accompanied by changes in total mRNA ( Figure 4B cf . npl3Δ and npl3Δbre1Δ , Exon feature ) . Because both Npl3 and Bre1 have been shown to have effects on transcription itself [13]–[16] , [66]–[68] , we normalized for changes in exon level by calculating an Intron Accumulation Index [69] ( see Materials and Methods ) for each intron-containing gene ( Figure S2 and Table S5 ) . The histogram of genes with an Intron Accumulation Index of greater than 0 . 3 ( Figure 4C ) , shows that even when normalized for changes in transcript levels , the total number of genes with a splicing defect , as well as the severity of the defect , is increased in the npl3Δbre1Δ strain as compared to npl3Δ alone . We also found that in the presence of wild-type Npl3 , a strain lacking BRE1 has a mild but reproducible splicing defect ( Figure 4B , bre1Δ –shown is an average of 5 biological replicates ) . While the majority of pre-mRNAs are not affected by the deletion of BRE1 , a small subset of pre-mRNAs accumulates in bre1Δ at 37°C ( Figure 4B , e . g . , DBP2 , LSB3 , YOP1 ) . This suggests that Bre1 has a role in pre-mRNA splicing , independent of the sensitivity caused when NPL3 is deleted . This finding was confirmed when we calculated Intron Accumulation Indices for a strain lacking BRE1: a small number of genes exhibit defective splicing in the bre1Δ strain ( Figure 4C and Figure S2 ) . We validated these splicing defects for several genes using a qPCR assay ( Figure S3 ) . The lack of a significant growth defect in the bre1Δ strain ( Figure 1E ) is consistent with the idea that yeast can tolerate a modest splicing defect at a small number of non-RPGs . If the splicing defect exacerbation we observed with npl3Δbre1Δ was due to loss of H2B ubiquitination , we would then expect this exacerbation to be phenocopied by a strain with the H2B lysine to arginine point mutant used earlier ( Figure 1E ) . We did , in fact , find that the htb1K123R point mutation exacerbated the splicing defect observed in the npl3Δ mutant at many genes ( Figure 4B ) , further implicating the ubiquitination of H2B in splicing . This is also evident when normalizing for the changes in exon levels in the npl3Δhtb1K123R strain ( Figure 4C ) . In plotting the Intron Accumulation Index values of this strain , we find that the subset of affected genes overlaps extensively with the subset of genes affected in the npl3Δbre1Δ double mutant ( Figure S2 ) . We have shown that deletion of UBP8 partially suppresses the npl3Δ growth defect , and this is most pronounced at 16°C ( Figure 2B ) . We therefore tested whether deleting UBP8 would suppress the splicing defect of a strain lacking Npl3 , as predicted by the genetic interaction . Surprisingly , deletion of UBP8 instead exacerbated the splicing defect observed in the npl3Δ strain ( Figure 4B cf . npl3Δ and npl3Δubp8Δ ) , implying that the growth suppression is related to some other function of Npl3 . Notably , however , these microarray results indicate that in the absence of Npl3 , the complete cycle of H2B ubiquitination and de-ubiquitination is required for efficient splicing . To begin to investigate how Bre1 affects splicing , we used chromatin immunoprecipitation ( ChIP ) to test the prediction that Bre1 is required for association of the splicing machinery . However , we did not observe a significant Bre1-dependent decrease in U1 ( Prp42 ) , Mud2 , or U2 ( Lea1 ) association with genes whose splicing was inhibited in bre1Δ or npl3Δbre1Δ strains ( data not shown ) , suggesting an alternative mechanism by which Bre1 modulates splicing ( see discussion ) . In light of our data showing that a bre1Δ strain exhibited a mild splicing defect , we carried out directed genetic analyses to test for interactions between BRE1 and genes encoding other splicing factors , particularly those that genetically interact with Npl3 [7] . Just like a deletion of NPL3 , deleting BRE1 caused synthetic sickness when combined with deletion of NAM8 ( U1 snRNP ) , MUD2 , LEA1 ( U2 ) , or SNU66 ( U5 ) , further connecting Bre1 functionally with splicing ( Figure 5A ) . Interestingly , the growth of the bre1Δ strain was also compromised by deletion of the U2 snRNP component CUS2 , which does not genetically interact with npl3Δ [7] . Thus , although we approached these experiments through the lens of Npl3 , these genetic observations provide further support that Bre1 has independent interactions with the splicing machinery . Consistent with a lack of splicing defect upon UPB8 deletion , we and others generally did not observe genetic interactions between UBP8 and early or late splicing factors ( Figure 5B and [70] ) . There is one notable exception however; deletion of UBP8 suppressed the snu66Δ cold-sensitive growth defect ( Figure 5B ) . Taken together , these data highlight the fact that the H2B ubiquitination pathway is linked to splicing , even in the presence of wild-type Npl3 .
Our genetic screens revealed that a number of Npl3 genetic interactions center on the histone H2B ubiquitination cycle . Specifically , mutant strains that lack wild-type levels of ubiquitinated H2B ( rad6Δ , bre1Δ , lge1Δ , htb1K123R , paf1Δ , cdc73Δ , and leo1Δ ) exacerbate the growth defect of an npl3Δ strain at all temperatures tested ( Figure 1 ) . We also observed a physical interaction between Npl3 and Bre1 by co-immunoprecipitation ( Figure 3 ) and showed that the splicing defect caused by deletion of NPL3 is exacerbated by the additional deletion of BRE1 or mutation of H2B ( htb1K123R ) , thus implicating H2B lysine 123 mono-ubiquitination in splicing ( Figure 4 ) . We previously demonstrated that Npl3 primarily affects the splicing of RPGs [7]; here , we see that in the sensitized background of a strain in which RPG splicing is made limiting ( npl3Δ ) , the histone H2B ubiquitination cycle is an important contributor to RPG splicing . Recent studies have shown that deletion of components of the cap-binding complex ( CBC ) or commitment complex causes defective splicing of the SUS1 pre-mRNA [70] , [74] . Sus1 is a recently discovered component of the histone de-ubiquitination module of SAGA [75] and if the SUS1 transcript is not properly spliced , it leads to elevated levels of ubiquitinated H2B . Given the physical [29] , [76] and genetic ( Table S1 and [76] ) connections between Npl3 and the CBC , we sought to determine whether Npl3 also affects SUS1 splicing and , therefore , H2B ubiquitination . However , Hossain , et al . have recently shown that deletion of NPL3 has no effect on SUS1 splicing [74] , a result we independently confirmed in our npl3Δ strain ( Figure S4 ) . Furthermore , we extended this analysis and determined that , unlike in cbcΔ strains , global levels of ubiquitinated histone H2B are not discernibly altered in the npl3Δ strain ( Figure S1 ) . While we cannot rule out a change in the dynamics of the ubiquitination cycle or gene-specific effects , our microarray results support a model in which the full histone ubiquitination cycle promotes RPG splicing , a process that becomes critical in the absence of NPL3 . Along these lines , it is noteworthy that data from Schulze et al . and Shieh et al . have revealed that chromatin over these genes is enriched for ubiquitinated H2B [77] , [78] . We also identified suppressive genetic interactions between NPL3 and genes responsible for removal of ubiquitin from H2B ( ubp8Δ , sgf11Δ , and sgf73Δ ) , suggesting that H2B de-ubiquitination is also linked to Npl3 function . Surprisingly , however , deletion of UBP8 did not suppress the splicing defect in npl3Δ , but rather exacerbated it ( Figure 4 ) . Thus , it seems the positive genetic interaction may be due to Ubp8 involvement in a splicing-independent function of Npl3 . The exacerbation seen in the microarray experiments shows that both halves of the cycle of H2B ubiquitination and de-ubiquitination are required for optimal splicing , as is the case for transcriptional activation [23] . Likewise , both halves of the H3 acetylation and deacetylation cycle , performed by Gcn5 and Hos2/3 , respectively , promote spliceosome assembly at the ECM33 gene [43] , [79] . Thus , these two examples point to a general function of dynamic histone modification cycles in maintaining fine control over co-transcriptional splicing , and may explain the synthetic lethality we observed between NPL3 and the acetylation module of the SAGA Complex ( Figure 1C and Table S1 ) . We found that even in the presence of wild-type NPL3 , Bre1 has genetic connections to the splicing machinery as a whole . Specifically , we found that deletion of BRE1 causes growth defects in early and late splicing factor deletion backgrounds ( particularly at extreme temperatures; Figure 5 , 16°C and 37°C ) , which alone show little to no growth defect . These negative genetic interactions can indicate two alternative but not mutually exclusive models for a functional relationship between the H2B ubiquitination and splicing machineries . One model is based on the fact that deletion of specific splicing factors is known to increase the levels of ubiquitinated H2B [74] , a phenotype that should be relieved by deletion of BRE1 , the sole H2B ubiquitin ligase [19] , [20] . Because this model predicts an epistatic or positive genetic interaction between BRE1 and the genes that encode splicing factors , the negative genetic interactions that we actually observe ( Figure 5 and [70] ) require an alternative model , perhaps one in which the growth defects are due to poorer overall splicing efficiency in these strains . Indeed , deletion of BRE1 alone caused a modest but reproducible splicing defect , seen in the microarray in Figure 4 . A large fraction of Bre1-dependent splicing events involve non-RPGs , and thus define a distinct role for Bre1 in splicing , apart from Npl3 . Shieh et al . [78] recently found that the pattern of this modified histone at non-RPGs shows a remarkable demarcation of intron/exon structure: low levels in the intron , followed by a marked increase at the intron – exon boundary . While the functional significance of this pattern of H2B ubiquitination is unknown , we propose that it may be relevant for the splicing of non-RPGs , as gauged by the splicing defect in a strain that no longer has this mark . We note that the single mutant htb1K123R has a milder splicing defect than the bre1Δ strain ( Figure 4C and Figure S2 ) , opening the possibility of an additional role of Bre1 in splicing that is independent of H2B ubiquitination . Indeed , Bre1-dependent ubiquitination of Swd2 , a protein in both COMPASS and the Cleavage and Polyadenylation Stimulatory Factor complex [46] , has been shown to regulate mRNA export from the nucleus [80] . Npl3 has previously been implicated in mRNA export [17] , [18] , [27] , [28] in a strain background where Npl3 is an essential protein . However , our data argue against the possibility that the genetic interactions we observed here are due to an adverse effect on mRNA export . In the present strain background ( S288C ) , in which Npl3 is non-essential , the npl3Δ strain does not exhibit the nuclear localization of bulk poly-adenylated mRNA characteristic of an export defect ( Figure S5 ) ; nor does further deletion of BRE1 in an npl3Δ strain cause an export defect ( Figure S5 ) . Furthermore , we found that the npl3S411A phosphorylation mutation , which blocks 3′ end formation [13] , [14] and mRNA export [17] , does not cause a block in pre-mRNA splicing ( Figure S6 ) . This argues against the reported splicing defects being the indirect result of feedback from these downstream defects in mRNA processing . We tested the prediction that Bre1 , like Npl3 , promotes spliceosome recruitment , but found that deletion of BRE1 did not affect the association of U1 ( Prp42 ) , Mud2 , or U2 ( Lea1 ) with chromatin at genes whose splicing is dependent on Bre1 ( data not shown ) . It may be that H2B ubiquitination is required for the recruitment of a later splicing factor or , as the H2B ubiquitination cycle regulates PolII passage through a gene [66] , [81] , it is possible that disruption of this cycle causes a subtle alteration of spliceosome dynamics that is not observable by ChIP . Furthermore , we cannot rule out the possibility that Bre1 has a ubiquitination target within the spliceosome or even ubiquitinates Npl3 itself . Both splicing and mRNA processing are largely co-transcriptional processes in eukaryotes , from yeast [2] to human [82]–[84] . Our survey of NPL3 genetic interactions has revealed a multitude of chromatin-connected factors with potential links to splicing and mRNA processing; overall , these results are thus consistent with an “integrator” role for Npl3 in gene expression ( Figure 6 ) . Our data provide a basis for the further study of the coupling of SR/hnRNP-dependent mRNA processing and transcription within a chromatin context , and have led to the discovery of Npl3-dependent and independent roles for Bre1 and histone H2B ubiquitination in splicing .
Unless otherwise indicated , yeast were grown as described in [85] . The npl3Δ::NatNT2 “magic marker” query strain used in the SGA was YTK232D , and was previously used in [26] . YTK232D was generated using techniques outlined in [86] . Briefly , the NPL3 open reading frame was replaced with NatNT2 via integration of a PCR product generated with primers ( 5′- TACTTTTGAAGGAATCAAAATTAAGCAATTACGCTAAAACCATAAGGATAACATGGAGGCCCAGAATACCC-3′ ) and ( 5′-GTTTTAAAACAATTCATATCTTTTGTTAATTTCTCCTTTTTTTTTCTCAACCAGTATAGCGACCAGCATTC-3′ ) into the SGA diploid strain [87] . The diploid was sporulated and the MATα npl3Δ::NatNT2 query strain was isolated by tetrad dissection , followed by re-selection of magic markers on SD medium lacking leucine and arginine but containing canavanine , s-AEC , and clonNAT [SD - LEU/ARG+100 µg/mL canavanine+100 µg/mL S- ( 2-Aminoethyl ) -L-cysteine hydrochloride+100 µg/mL clonNAT] . The NPL3 deletion was confirmed by PCR , and by Western blot for the absence of Npl3 using an α-Npl3 antibody [63] . The Synthetic Genetic Array was performed as described in [24] with the following exceptions: Here the npl3Δ query strain ( YTK232D ) was mated to the MATa KanMX-marked deletion collection ( OpenBiosystems: www . openbiosystems . com; formerly Research Genetics , Huntsville , AL ) . The collection was arrayed in duplicate in 384-well colony format using automated pinning ( Colony Arrayer ) and grown at 30°C for 2 days . Mating was carried out at 30°C for 2 days . Sporulation was carried out at 30°C for 7 days . MATa double mutants were selected on SD medium lacking histidine and arginine but containing canavanine , S-AEC , G418 , and clonNAT [SD - HIS/ARG+100 µg/mL canavanine+100 µg/mL S- ( 2-Aminoethyl ) -L-cysteine hydrochloride+150 µg/mL G418 and 100 µg/mL clonNAT] . Double mutant arrays were re-pinned in replicate and photographed after the following incubations: 30°C for 5 days , 37°C for 5 days , or at 16°C for 4 and again after 8 days . Photographs were visually inspected for growth at 16°C ( to identify suppressive interactions ) or lack of colony growth at 30°C or 37°C ( to identify synthetic lethal interactions ) . The npl3Δ::NatNT2 strain used for directed genetics ( YTK234D ) was previously used in [7] . Unless otherwise indicated , YTK234D was crossed to a series of MATa KanMX4-marked deletion strains; diploids were selected by plating on YPD plates+100 µg/mL clonNAT+150 µg/mL G418 . Double mutants were isolated by tetrad dissection or random sporulation , as indicated in Table S6 . All single mutants were validated by PCR for the knockout chromosome prior to crossing to YTK234D . The HTB1-WT ( WHY334 ) and htb1-K123R ( WHY326 ) strains contain htb2Δ::HygX4l and the indicated htb1 allele as the sole copy of H2B ( gifts from W . Hwang and H . Madhani ) . They were mated as above , except the diploid strains were selected on YPD+100 µg/mL hygromycin+100 µg/mL clonNAT . Because the htb1 allele is unmarked , the final npl3ΔHTB1 and npl3Δhtb1K123R strains were confirmed by sequencing the HTB1 gene and Western blot for the Npl3 protein . Genetic interactions with the bre1H665A allele were analyzed using a set of plasmids provided by the Shilatifard lab [19] , designed to complement a bre1Δ allele . Complementation was achieved by plasmid transformation into YM1740 ( bre1Δ ) or YTK391B ( npl3Δbre1Δ ) , which were maintained on SD -LEU plates . The bre1Δ::NatNT2 ( EMy32 ) and ubp8Δ::NatNT2 ( EMy442 ) strains were created by replacement of the endogenous ORF with NatNT2 , as described in [86] . These strains were subsequently mated to nam8Δ , mud1Δ , mud2Δ , syf2Δ , and snu66Δ ( for bre1Δ ) , and lea1Δ ( for bre1Δ and ubp8Δ ) from the deletion collection and double mutants were isolated via tetrad dissection . For the rest of the ubp8Δ genetics , the ubp8Δ::KanMX4 strain from the deletion collection was mated to MATα NAT-marked “magic marked” splicing factor deletion strains . These splicing factor deletion strains were made by replacing the KanMX-marked ORFs with NatNT2 , followed by crossing to a “magic marked” wild-type ( YTK609 ) to isolate “magic marked” NAT-marked MATα spores . The ubp8Δ::KanMX4 strain was mated to each NAT-marked splicing factor deletion strain and MATa double mutants were isolated by tetrad dissection followed by selection on SD medium lacking histidine and arginine but containing canavanine , S-AEC , G418 , and clonNAT [SD - HIS/ARG+100 µg/mL canavanine+100 µg/mL S- ( 2-Aminoethyl ) -L-cysteine hydrochloride+150 µg/mL G418 and 100 µg/mL clonNAT] . For individual growth assays , log-phase yeast were diluted to OD600 = 0 . 1 , spotted onto YPD plates ( unless specifically mentioned ) in a 5-fold dilution series and grown at the indicated temperatures . For each cross , growth of the double mutant was confirmed for ≥2 double mutant isolates , and a representative isolate is shown . The single mutants and wild-type strains shown are either parental strains , or were re-isolated from tetra-type tetrads . The bre1H665A and BRE1 strains were serially diluted onto SD –LEU plates . We sought to integrate the diverse sources of genetic interaction information available to us in order to create a comprehensive dataset for statistical analyses . Because the stronger synthetic interactions were identified in the 30°C SGA , we began with this list of genes whose deletion caused lethality in combination with npl3Δ ( see Table S1 – 30°C ) and added genes identified as causing markedly decreased growth , as gauged by serial dilution , or lethality , as gauged by loss of double mutant spore after tetrad dissection ( Figure 1C , 1D , 1E and Table S1 ) . We further added to this list genes identified as synthetic sick or lethal in the E-MAP [26] i . e . , having a genetic interaction score of ≤−2 . 5 . Biological process definitions were obtained from the Gene Ontology annotations maintained at SGD [88] on April 15th 2012 . Forty-five high-level ( GO Slim ) terms were used and are included in Table S7 . Protein complex definitions were obtained from a manually curated list , CYC2008 [89] , and augmented with the RAD6 Complex ( RAD6 , BRE1 , LGE1 ) , which was not annotated when the list was created . A hypergeometric test was used to identify complexes and processes that were significantly enriched with genetic interactions . Complex enrichment p-values were corrected for multiple testing using the empirical re-sampling method of Berriz et al . [90] ( as 409 complexes were assessed for enrichment ) , while process enrichment p-values were corrected for using the simpler Bonferoni correction . The results of the these analyses are included in Table S2 ( by process ) and Table S3 ( by complex ) . The network diagram in Figure 1B was drawn using Cytoscape [91] . For Figure 1B , complexes were referred to by their more common names . The Figure 2A diagram was created to highlight a subset of suppressive interactions identified in the 16°C SGA and the full list of suppressors is available in Table S4 . Co-immunoprecipitation assays were performed as in [7] with extracts from the indicated GFP-tagged or HA-tagged strains . The Nup188-HA strain contains a plasmid encoding Nup188-3XHA . The other strains were tagged endogenously . Briefly , samples were separated by 10% SDS-PAGE and probed by Western blot with either monoclonal α-GFP ( Roche 1814460 ) , α-HA ( 12CA5; Roche 11583816001 ) , or polyclonal α-Npl3 antibodies [63] . Total samples equivalent to 1/60th of the input were analyzed in parallel . Cultures were grown according to standard techniques [85] in rich medium supplemented with 2% glucose . Strains were cultured overnight to saturation and diluted to OD600 = 0 . 1 in the morning . The strains were allowed to grow at 30°C until reaching mid-log phase ( OD600 = 0 . 5–0 . 7 ) , at which point they were collected ( for Figure S6 ) , or rapidly shifted to either 37°C for 30 minutes or 16°C for 2 . 5 hours , as indicated . Cultures were collected by centrifugation and snap frozen in liquid nitrogen . Total cellular RNA was isolated using hot acid phenol followed by isopropanol precipitation , as outlined in [92] but with modifications detailed in [93] . cDNA from each strain was synthesized , and labeled with Cy3 or Cy5 according to the low-throughput sample preparation method described in [65] . The optimized oligos listed in [65] were robotically arrayed onto poly-L-lysine coated glass slides ( slides from ThermoScientific C40-5257-M20 ) and slides were processed using the protocols detailed in [65] , [94] Each biological replicate contains 6 technical replicates for each feature as well as dye-flipped replicates . Microarrays were scanned using Axon Instruments GenePix 4000B at 635 nm and 532 nm wavelengths and image analysis was done using Axon Instruments GenePix Pro version 5 . 1 . Spots were manually removed from analysis if they contained obvious defects or uncharacteristically high background; the ratio of the median intensities for 535 nm and 625 nm was calculated for each remaining spot . Technical replicate spots and dye flipped replicates were combined and normalized as in [65] . The resulting log2-transformed values for each feature were averaged over 2–5 biological replicates . Averaged data were subjected to hierarchical clustering using average linkage , and uncentered Pearson correlation as the similarity metric using Cluster 3 . 0 [95] . Resulting heat maps in Figure 4 , Figure S2 , and Figure S6 were created using Java Treeview [96] . To normalize for changes in total expression evident in the microarrays , Intron Accumulation Indices ( IAI ) were calculated for each intron containing gene as in [69]; specifically , we calculated log2 ( Intronmutant/IntronWT ) -log2 ( Exonmutant/ExonWT ) for each gene . The IAI heat map is shown in Figure S2 . These values were converted into a histogram for Figure 4 using the following cutoffs: −0 . 3≥IAI≥0 . 3 . The dT50 assay was performed based on the protocol outlined in [97] with the following modifications . Specifically , 2 mL cultures were fixed in 5% formaldehyde for 1 . 5 hours after having reached OD600 = 0 . 2–0 . 3 . Cells were washed 4 times in wash buffer ( 100 mM Potassium Phosphate , 1 . 2 M Sorbitol ) before a 40-minute treatment with 27 µg zymolyase at 37°C . An additional fixation was performed in 8% paraformaldehyde in PBS+10 mM MgCl2 and spheroplasted cells were applied to poly-lysine-treated chamber slides ( LabTek 178599 ) . Attached cells were treated with ice-cold methanol ( −20°C ) and allowed to dry . Hybridization to digoxin-conjugated dT50 oligo in blocking buffer was performed at 37°C overnight . Chambers were washed with 2× ( 20 minutes ) , 1× ( 20 minutes ) and 0 . 5× SSC ( 10 minutes at 37°C ) before a 30-minute incubation with FITC- conjugated anti-Digoxin Fab fragments ( Roche 1207741 ) in blocking buffer ( 1∶25 dilution , 37°C ) . Antibody was aspirated and three 5-minute washes of PBS +10 mM MgCl2 were performed . Chambers were treated with 0 . 5 mg/mL DAPI for 2 minutes and slides were mounted using ProLong Gold Antifade Reagent ( Invitrogen P36934 ) according to manufacturer instructions . Slides were visualized using an Olympus BX60 microscope equipped with FITC HiQ and DAPI HiQ Filters ( Chroma Technology Corporation ) . The assay was performed on two biological replicates and representative images are shown . Specificity of the probe and FITC labeling was determined by incubation with hybridization mix lacking probe ( data not shown ) . SUS1 splicing efficiency was measured essentially according to the non-radioactive protocol described in [74] . Specifically , 10 µg RNA from cultures grown at 30°C was treated with DNaseI ( Promega ) and RNA was converted to cDNA using 1 µg SUS1 Reverse primer [70] . cDNAs were diluted 1∶200 and 10 µL was used in a 25 µL PCR ( BioRad iProof ) with SUS1-specific primers [70] . 25 cycles of PCR were performed and the resulting products were separated on an 8% polyacrylamide gel . Gels were stained using SybrGold and bands were quantified using an AlphaImager HP camera and software . 2–3 technical replicates of 2 biological samples were performed . Shown are a representative gel and the average and standard deviations of all technical replicates . A no-Reverse Transcriptase control was performed for each sample and none showed amplification ( data not shown ) . A TCA precipitation was performed on strains grown at 30°C [98] and samples were run on a 15% SDS polyacrylamide gel and transferred to PVDF membrane . Membrane was blocked using Li-Cor blocking buffer , followed by incubation of a 1∶1000 dilution of α-H2B antibody ( Active Motif 39237 ) overnight at 4°C . Visualization of bands was achieved with a secondary antibody conjugated to infrared dye ( LI-COR 926-32211 ) . The membrane was scanned using the LI-COR Odyssey scanner and software . Shown is a representative Western blot . The assay was performed with 3 biological replicates and shown are the average and standard deviation of the three replicates . RNA was extracted as described above from strains grown under the same conditions as for the microarray experiment . Five µg RNA were treated with DNaseI ( Promega ) before being primed with random 9-mers and reverse transcribed . Samples were diluted as necessary and 10 µL were used in each qPCR . qPCRs were run on a C1000 ThermoCycler ( BioRad ) with an annealing temperature of 55°C . Each qPCR run was finished with a melt curve to determine the homogeneity of the amplified product . Starting quantity was calculated using a standard curve for each primer set . 2–4 technical replicates were performed for 1–5 biological replicates . Error bars represent standard deviation for biological replicates . For samples with 1 biological replicate , standard deviation of technical replicates is shown with uncapped error bars ( Figure S3 ) . A no-Reverse Transcriptase control was also generated for each RNA sample and these samples yielded negligible amplification ( data not shown ) . Primers used in the qPCR are listed in Table S8 . Each gene was measured using intron- and exon- specific primer sets . The Intron/Exon ratio for each mutant was normalized to its corresponding wild-type before averaging . | Pre-messenger RNA splicing is the process by which an intron is identified and removed from a transcript and the protein-coding exons are ligated together . It is carried out by the spliceosome , a large and dynamic molecular machine that catalyzes the splicing reaction . It is now apparent that most splicing occurs while the transcript is still engaged with RNA polymerase , implying that the biologically relevant splicing substrate is chromatin-associated . Here , we used a genetic approach to understand which factors participate in the coordination of transcription and splicing . Having recently shown that the Npl3 protein is involved in the recruitment of splicing factors to chromatin-associated transcripts , we performed a systematic screen for genetically interacting factors . Interestingly , we identified factors that influence the ubiquitin modification of histone H2B , a mark involved in transcription initiation and elongation . We show that disruption of the H2B ubiquitination/de-ubiquitination cycle results in defects in splicing , particularly in the absence of Npl3 . Furthermore , the ubiquitin ligase , Bre1 , shows genetic interactions with other , more canonical spliceosomal factors . Taken together with the myriad Npl3 interaction partners we found , our data suggest an extensive cross-talk between the spliceosome and chromatin . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | The Yeast SR-Like Protein Npl3 Links Chromatin Modification to mRNA Processing |
Glutamate-gated chloride channel receptors ( GluClRs ) mediate inhibitory neurotransmission at invertebrate synapses and are primary targets of parasites that impact drastically on agriculture and human health . Ivermectin ( IVM ) is a broad-spectrum pesticide that binds and potentiates GluClR activity . Resistance to IVM is a major economic and health concern , but the molecular and synaptic mechanisms of resistance are ill-defined . Here we focus on GluClRs of the agricultural endoparasite , Haemonchus contortus . We demonstrate that IVM potentiates inhibitory input by inducing a tonic current that plateaus over 15 minutes and by enhancing post-synaptic current peak amplitude and decay times . We further demonstrate that IVM greatly enhances the active durations of single receptors . These effects are greatly attenuated when endogenous IVM-insensitive subunits are incorporated into GluClRs , suggesting a mechanism of IVM resistance that does not affect glutamate sensitivity . We discovered functional groups of IVM that contribute to tuning its potency at different isoforms and show that the dominant mode of access of IVM is via the cell membrane to the receptor .
Glutamate-gated chloride channel receptors ( GluClRs ) are a major class of pentameric ligand gated ion channels ( pLGICs ) [1] that mediate neuronal and muscular inhibition [2 , 3] . They are expressed exclusively in invertebrates , making them ideal targets for the development of pesticides with minimal risk of unwanted activity at vertebrate pLGICs . GluClRs are of interest because they are an excellent model for understanding pLGIC structure at high resolution [4 , 5] , function and pharmacology [6 , 7] , and because many invertebrates that express them are commercially important pest species in agriculture [8–10] , aquaculture [11] and in veterinary and human health [12] . Ivermectin ( IVM ) is a highly effective , broad-spectrum anthelminthic drug that targets GluClRs . Its binding site is formed between the first ( M1 ) and third ( M3 ) transmembrane domains contributed by adjacent subunits [5] ( Fig 1A and 1B ) . The drug is used extensively as an insecticide and anti-parasitic agent [13] , where its mechanisms of action include paralysis of feeding structures , such as the pharynx , in adult organisms [14–16] and impairment of larval viability [13 , 17] . However , resistance to IVM and related macrocyclic lactones , such as abamectin , is an emerging concern that threatens food production [9 , 18] as well as human [19] and animal health [20] on a global scale . Underscoring the imperative for a greater understanding of the molecular targets of pesticides and mechanisms of resistance , studies predict that increasing seasonal temperatures due to climate warming will significantly exacerbate agricultural loss that results from an increase in populations of herbivorous pests [10 , 21] and veterinary parasites [9] . The chemical synthesis of new IVM derivatives is a key strategy in overcoming IVM resistance as it has the potential to probe specific molecular groups of IVM [22] and their interactions with IVM sensitive and resistant forms of GluClR . Wild , IVM-resistant pest isolates that express missense point mutations to GluClR subunits have been identified in the herbivorous species , Plutella xylostella [23 , 24] ( A30’V ) and Tetranychus urticae [25–27] ( G36’D , G36’E ) and in a laboratory-induced mutation to a GluClR subunit of Drosophila melanogaster ( P23’S ) upon exposure to IVM [28] ( Fig 1B ) . Point mutations to GluClR subunits are also being reported in human head lice , Pediculus humanus , that include the A55’V , and S46P and H272R , the latter two mutations being located in the extracellular and intracellular domains of the receptor , respectively [29] . In addition to mutations , the sensitivity of GluClRs to IVM is affected by subunit composition . Cully et al . 1994 first identified GluClR subunit isoforms in the nematode , Caenorhabditis elegans and demonstrated that they could assemble as homo- and heteromeric receptors with different pharmacological properties [1] . C . elegans expresses a β subunit that is insensitive to IVM when expressed as a homopentamer , and acts to reduce IVM sensitivity when co-assembled with the IVM-sensitive α subunit [1] ( Fig 1B ) . A naturally IVM-resistant variant of an α subunit isoform has also been discovered in a strain of C . elegans . IVM insensitivity can be overcome by overexpressing IVM-sensitive subunits in the resistant C . elegans strain [30] . Moreover , IVM effectively kills the larvae of the human pathogens , Onchocerca volvulus and Wuchereria bancrofti , but is less effective against the adult stages of these organisms [13] , suggesting a developmental switch in GluClR subunit composition . Evidence for selection pressure on GluClR isoform expression that confers IVM insensitivity comes from IVM-resistant isolates of the bovine pathogens , Cooperia oncophora and Ostertagia ostertagi , which demonstrate transcriptional downregulation of genes that encode IVM-sensitive GluClR subunits [31] . H . contortus is one of the world’s most economically important agricultural parasites that infects domestic ruminant animals , and field resistance to IVM is well-documented [32] . It has also been reported to infect wild ruminants , including species that are critically endangered [33] . This endoparasitic nematode has six GluClR genes encoding at least eight subunits [34] . Functional expression of wild-type homomeric GluClRs comprising α ( avr-14b ) subunits have been reported previously , including mutations to this subunit that greatly reduce IVM sensitivity [6 , 35 , 36] . Although the β subunit of H . contortus was shown to have overlapping distribution patterns with the α ( avr-14b ) and other α subunit isoforms [14] , little is known about its function . Notably , it is not known if the β subunit forms functional homomeric receptors or whether it can combine with other subunits to form heteromers . It also remains to be determined which homomeric or heteromeric combinations of α and β subunits cluster at postsynaptic sites to mediate inhibitory postsynaptic currents ( IPSCs ) . This deficiency represents an obstacle to understanding basic invertebrate neurobiology and determining the effects of drugs on neuronal and muscular inhibitory input , including highly lipophilic drugs , such as IVM that exhibit quasi-irreversible effects on GluClRs [37] . Due to its lipophilic nature , IVM is believed to diffuse through the body cuticle of nematodes [38] and partition into cell membranes of target organisms where it reaches a high local concentration [39] . We have recently shown that single receptor active periods of homomeric α ( avr-14b ) GluClRs ( α GluClRs ) increase in duration over 1–2 minutes after exposure to IVM , suggesting that the drug equilibrates in the membrane over time to produce maximum receptor potentiation [35] . In this study , our aims were to determine ( 1 ) whether the β subunit of H . contortus can assemble as both β homomeric and αβ heteromeric GluClRs ( β GluClRs and αβ GluClRs ) , ( 2 ) whether subunit composition determines IVM sensitivity , ( 3 ) the potency of synthetically modified analogues of IVM ( 4 ) whether β and αβ GluClRs mediate IPSCs with different properties , and finally , ( 5 ) whether the membrane partitioning and diffusion properties of IVM correspond to the time-course of current potentiation .
To test for functional expression of β and αβ GluClRs , cDNA encoding the β subunit was injected into oocytes either alone or with cDNA encoding the α subunit , at a ratio of ( α:β ) 1:1 or 1:50 . Example glutamate-gated currents obtained from oocytes injected with the β subunit alone and the α and β subunits at a ratio of 1:50 are shown in Fig 1C . These data clearly demonstrate that the β subunit can indeed form functional β GluClRs and , given the differential glutamate sensitivity of co-injected oocytes , can assemble with the α subunit to form αβ GluClRs . The maximal currents for the α , αβ and β GluClRs were ( in μA ) 3 . 7 ± 0 . 8 , 2 . 0 ± 0 . 4 ( 1:1 ) , 2 . 2 ± 0 . 4 ( 1:50 ) and 1 . 9 ± 0 . 3 , respectively and were not statistically different . Complete glutamate concentration-response experiments were done in oocytes injected with the either cDNA encoding the α or β subunit and in oocytes co-injected with both cDNAs at ratios of 1:1 and 1:50 ( Fig 1D ) . These data show the β GluClRs are substantially less sensitive to glutamate with an EC50 of 394 μM compared to α homomers and the αβ heteromers ( p < 0 . 001 ) , all of which exhibited similar glutamate sensitivities , with EC50s of 28 μM ( α homomers ) , 40 μM ( αβ , 1:1 ) and 44 μM ( αβ , 1:50 ) ( Table 1 ) . Notably , the fitted concentration-response plots for both heteromeric combinations of receptors did not exhibit inflections typical of mixtures of distinct receptor stoichiometries with differential EC50s , as has been shown for GABA-gated pLGICs [40] . These data suggest that co-injected oocytes express mostly homogeneous populations of heteromeric receptors , and that a given injection ratio gives rise to a particular homogeneous stoichiometry . IVM concentration-response experiments were carried out to obtain additional evidence of populations of receptors that were a function of the oocyte injection ratio . Example IVM-induced currents for α GluClRs and αβ GluClRs at both injection ratios are shown in Fig 2A–2C . The corresponding group concentration-response data for four combinations of GluClRs are shown in Fig 2D . These plots indicate that α GluClRs are highly sensitive to IVM with an EC50 of 22 nM and a maximal response equivalent to 59% of that seen with saturating glutamate . Conversely , β GluClRs were effectively insensitive to IVM with an EC50 > 10 μm . The heteromeric combinations exhibited intermediate IVM sensitivities with the 1:1 injected oocytes being more sensitive ( EC50 of 86 nM , 26% of saturating glutamate response ) than the 1:50 injected oocytes ( EC50 of 141 nM , 33% of saturating glutamate response ) . The % of saturating glutamate response was not significantly different between the two injection ratios . However , the EC50s for the two heteromeric combinations were statistically different from the α GluClRs and from each other ( Table 1 ) . Together these data demonstrate that β subunits of H . contortus and C . elegans [1] exhibit some functional similarities . They both form homomeric receptors that are insensitive to IVM and can combine with α subunits to reduce IVM sensitivity . However , unlike the β GluClRs of C . elegans , those of H . contortus are responsive to glutamate , albeit with a reduced sensitivity . The data also show that the β subunit can co-assemble with the α subunit without reducing the sensitivity of the receptors to the neurotransmitter , glutamate . Furthermore , IVM potency is reduced when oocytes are injected with an excess of β subunit cDNA , implying that GluClRs with a greater content of β subunit are less sensitive to IVM . This later inference also suggests that heteromeric αβ GluClRs of H . contortus can increase the proportion of β subunit as a function of its expression level . This is in contrast to the reported fixed stoichiometry of heteromeric GluClRs of C . elegans [41] . In our next series of experiments , we wished to see if modifying the IVM molecule might alter its potency differentially at α and αβ GluClRs . Three analogues were screened ( Fig 3A ) at α and αβ GluClRs ( 1:50 ) using a standard concentration of 30 nM ( Fig 3B ) . Changes to the IVM parent molecule were made on the basis of IVM and receptor or lipid membrane interactions as revealed by the crystallographic structure of the C . elegans α GluClR [5] . The maximum current elicited by each IVM analogue was normalised to that induced by a saturating concentration of glutamate ( 5 mM , Fig 3B ) and its efficacy ( maximal current ) was compared to that of IVM and between α and αβ GluClRs ( Fig 3C ) . The disaccharide moiety is predicted to protrude into the upper leaflet of the cell membrane ( Fig 1A ) and to form van der Waal ( VDW ) interactions with the loop that links the M2 and M3 domains of the α subunit of C . elegans [5] and H . contortus . Hydrolysis of IVM yielded the aglycone IVM-1 ( synthesis details are given in S1 File ) . IVM-1 retained differential efficacy between the two receptor isoforms but was significantly less efficacious than IVM only at α GluClRs ( p < 0 . 05 ) . Replacing the disaccharide with a methoxymethyl ether group ( MOMO ) produced IVM-2 ( S1 File ) . This derivative restored efficacy at α but not αβ GluClRs and abolished the differential sensitivity between the receptors . This result suggests that in addition to the predicted M2-M3 loop interactions , steric factors also contribute to the potentiating effects of IVM . IVM-2 ( and IVM and IVM-1 ) , with a β-configured 5-OH moiety , is predicted to form VDW interactions and hydrogen bonds with residues of the M2 and M1 domains [5] . The α-configured 5-OH epimer ( IVM-3 , S1 File ) , prepared from IVM-2 , produced the most marked reduction in potency ( p < 0 . 001 for both GluClRs ) relative to IVM . Moreover , this change also ablated the differential efficacy between the two GluClR isoforms ( Fig 3C ) . This result is in accord with the predicted role of the 5–OH moiety , which forms a hydrogen bond with the polar residues within the pore-lining M2 domain ( S15’ in C . elegans α [5] or S16’ and Q15’in H . contortus α and β , respectively ( Fig 1B ) . A previous study also found this position was important , as subtle changes to 5-O or 5-NOH resulted in dramatic decreases in activity when tested in an H . contortus larval assay [22] . Overall , differential efficacy between α and αβ GluClRs was maintained for IVM and IVM-1 whereas IVM-2 and IVM-3 were equally efficacious at both receptors . The preservation of efficacy of IVM-1 and IVM-2 compared to IVM at αβ GluClRs suggests that the β subunit likely contributes to IVM binding sites . The disaccharide and the 5-OH moiety of IVM are salient determinants of the potentiating potency of IVM . The data reveal two IVM-receptor interactions that accord well with predictions from structural studies [5] and suggest that with rational drug design it may be possible to develop IVM analogues to selectively target specific GluClR isoforms . Evidence suggesting the existence of glutamate-gated anion-selective post-synaptic receptors was first noted in arthropods over 40 years ago via observing membrane potential changes [2 , 3 , 42] . However , to our knowledge direct recordings of GluClR-mediated IPSCs have not been made in any invertebrate species . We have recently developed a heteroculture consisting of primary ( mammalian ) neurons and HEK293 cells transfected with defined post-synaptic receptors of vertebrates , along with the trans-synaptic protein neuroligin [43–45] . This heterosynapse preparation was used to examine if α and αβ GluClRs of H . contortus cluster at post-synaptic sites and respond to synaptically released glutamate . Cells transfected with the α ( Fig 4A ) or α and β ( 1:1 , Fig 4B ) subunits in co-culture with neurons exhibited prominent currents typical of those expressed at vertebrate synapses , consisting of a fast rise to peak followed by a slower , exponential decay back to baseline ( Fig 4C and 4D ) . These IPSCs were only observed in transfected cells and were blocked by picrotoxin ( Fig 4E ) , confirming that the receptors were anion-selective GluClRs [5 , 36] . These data clearly demonstrate that α and αβ GluClRs of H . contortus generate classic , picrotoxin-sensitive IPSCs . A clear , observable difference between IPSCs mediated by the two receptor isoforms was that αβ GluClRs decayed faster than those mediated by α GluClRs ( Fig 4C and 4D ) . The mean rise-times , peak amplitudes and decay times were measured and plotted as bar plots ( Fig 4F–4H ) . The group averages confirmed that only the decay times between the two receptor isoforms were significantly different , with those of the α GluClRs decaying with a time constant of 40 ± 2 ms ( n = 17 ) , whereas the decay time constant for the αβ GluClRs was 17 ± 2 ms ( n = 8 ) . The rise times were 4 . 6 ± 0 . 7 ms for the α GluClRs and 3 . 2 ± 0 . 3 ms for the αβ GluClRs . The peak amplitudes were the most variable of the three parameters , being 60 ± 9 pA and 34 ± 8 pA for the α and αβ GluClRs , respectively , likely reflecting variable expression levels of receptors . After validating the heterocultures as a reliable preparation for investigating IPSCs mediated by GluClRs , we made synaptic recordings while continuously applying IVM ( 5 nM ) to cells similarly transfected with α and αβ GluClRs . Two salient effects were observed in these recordings . Firstly , there was a steady downward deflection in baseline current over the course of the recording that was not different between cells expressing α ( Fig 5A ) and αβ ( 1:1 , Fig 5B ) GluClRs . These data were subsequently pooled for analysis of the tonic current component . The tonic currents plateaued at 17 ± 1 minutes ( n = 13 ) , had a fitted time constant of 483 ± 72 s and a mean amplitude of 304 ± 54 pA . Changes to the kinetic properties of IPSCs were also observed upon IVM application ( Fig 5C–5G ) . These were analysed 15 minutes after the commencement of the IVM application to ensure that the effects had equilibrated . For both receptor types , the decay and rise times and peak amplitudes increased significantly in the presence of IVM relative to IVM naïve cells . The decay time constant for α GluClRs slowed from 40 ± 2 ms to 100 ± 3 ms ( n = 6 ) and for αβ GluClRs from 17 ± 2 ms to 72 ± 4 ms ( n = 6 , Fig 5E ) . The increase in activation times were from 4 . 6 ± 0 . 7 ms to 19 ± 1 ms for α GluClRs and from 3 . 0 ± 0 . 3 ms to 10 ± 1 ms for the αβ heteromers ( Fig 5F ) . Again , the peak IPSC amplitudes were variable but increased significantly in the presence of IVM from 60 ± 9 pA to 122 ± 39 pA and from 34 ± 8 pA to 124 ± 18 pA for α and αβ GluClRs , respectively ( Fig 5G ) . A comparison of the magnitude changes of IPSC parameters was also made between the two receptor isoforms . Consistently greater decay and rise times were observed for the α GluClRs ( Fig 5E and 5F ) , suggesting that IPSCs mediated by these receptors produce greater inhibitory input than αβ GluClRs in the presence of IVM . The decay times of IPSCs mediated by αβ GluClRs were consistently and significantly faster than those of α GluClRs , in both the absence and presence of IVM . We previously showed that the decay times of macropatch currents mediated by α GluClRs incorporating an IVM-insensitive mutation ( G36’A ) were faster than those mediated by wild-type α GluClRs , and that this was due to briefer single receptor active periods and enhanced receptor desensitisation [35] . To explore whether a similar mechanism applies to β-containing GluClRs , we recorded single receptor currents in excised patches from cells transfected with either the α and β subunits ( 1:1 ) or the β subunit alone . These experiments were recorded in 2 μM and 3 mM glutamate as well as 2 μM glutamate plus 5 nM IVM at a clamped potential of ‐70 mV ( reversal potential = 4 . 0 mV , liquid junction potential = 4 . 7 mV and membrane potential = 78 . 7 mV ) [35] . Patches from cells transfected with both subunits revealed that the majority ( ~90% ) of single αβ receptors opened to an amplitude of 1 . 2 pA ( Fig 6A ) with the remainder opening to 0 . 7 pA ( Fig 6B ) . By contrast , patches from cells transfected with the β subunit alone invariably opened to an amplitude of 0 . 4 pA ( Fig 6C ) . Amplitude histograms confirmed two amplitude levels for αβ GluClRs and a single amplitude for the β GluClRs ( Fig 6D ) . Our data suggest that a 1:1 transfection ratio of α and β subunits results in a predominant stoichiometry , which exhibits an amplitude of 1 . 2 pA and a less frequent subunit combination that opens to an amplitude of 0 . 7 pA . We infer that the αβ GluClRs that open to 0 . 7 pA likely contain a greater proportion of β subunits . This is consistent with β GluClRs having the smallest amplitude and α GluClRs having the greatest amplitude ( 1 . 8 pA ) under similar recording conditions [35] . The calculated conductance values for the corresponding amplitudes for αβ GluClRs was 15 . 2 pS and 8 . 9 pS and for the β GluClRs was 5 . 1 pS . Single receptor active periods were analysed for mean duration and intra-activation open probability ( PO ) . For αβ GluClRs the activations from both stoichiometries were pooled to obtain mean active periods of 40 ± 10 ms ( n = 5 ) in 2 μM glutamate and 146 ± 15 ms ( n = 5 ) in 3 mM glutamate . Consistent with the example recordings presented in Fig 6A and 6B the respective POs were 0 . 54 ± 0 . 06 and 0 . 83 ± 0 . 03 in 2 μM and 3 mM glutamate ( Fig 6E ) . Current potentiation by IVM did not change the proportion of activations to 1 . 2 pA compared to 0 . 7 pA , but manifested as a marked increase in the mean duration of the active periods ( pooled , 876 ± 178 ms , n = 6 ) ( Fig 6A , 6B and 6E ) . This represents an over 20-fold increase compared to 2 μM glutamate alone and 6-fold increase compared to 3 mM glutamate . The PO in 2 μM glutamate plus 5 nM IVM was similar to that in 3 mM glutamate , being 0 . 83 ± 0 . 02 ( Fig 6E ) . β GluClRs exhibited the briefest active periods ( Fig 6C ) . At 2 μM glutamate , β GluClRs opened for a mean duration of 7 . 5 ± 1 . 8 ms and had a PO of 0 . 63 ± 0 . 07 ( n = 5 ) ( Fig 6F ) . The active durations increased in 3 mM glutamate to 164 ± 31 ms , whereas the PO remained unchanged ( 0 . 70 ± 0 . 11 , n = 5 ) ( Fig 6F ) . In contrast to the supersaturating effects IVM has at α and αβ GluClRs , it was ineffective at increasing the duration of active periods of β GluClRs beyond that measured in 3 mM glutamate . The active periods in 2 μM glutamate plus 5 nM IVM were 120 ± 21 ms in duration and the PO was 0 . 62 ± 0 . 04 ( n = 5 ) ( Fig 6F ) . Our single receptor measurements show that by increasing the active durations and PO of single GluClRs , IVM produces an increase in IPSC decay times and peak amplitude , respectively . Our data also demonstrate a correlation between single receptor active duration , IPSC decay times and IVM potentiation of IPSCs . Our IPSC and single channel data suggest that the potentiating effect of IVM at GluClRs requires several minutes to stabilise [35] . This relatively long delay could reflect the slow binding of IVM directly from the aqueous solution or slow IVM-induced conformational effects at GluClRs once bound [46 , 47] . Alternatively , it may reflect a slow accumulation rate of IVM into the membrane , possibly coupled with a slow lateral diffusion rate within the membrane that controls its access to receptor binding sites . To help distinguish between these possibilities , we synthesised a fluorescent IVM-bodipy probe ( IVM-bdpy , Fig 7A and S1 File ) to monitor the interaction of IVM with the cell membrane . This probe consisted of IVM-1 , which retains its activity at GluClRs ( Fig 3 ) , attached to a bodipy fluorophore . The resulting IVM-bdpy derivative exhibited spectral emission properties similar to free bdpy ( Fig 7B ) and was active at α GluClRs ( Fig 7C ) , albeit with a reduced potency ( Fig 7D ) . Two types of experiments were conducted with IVM-bdpy . The first was aimed at monitoring the time-course of membrane accumulation . Untransfected HEK293 cells were incubated in extracellular solution containing 500 nM IVM-bdpy and time-lapse images were taken over a period of 30 min ( Fig 8A ) . The change in total membrane fluorescence averaged from 5 repeat experiments was then plotted against time and fitted to a standard exponential function ( Fig 8B ) . In a control experiment , the corresponding accumulation rate was also determined for bdpy . To obtain the net membrane partitioning time-course for IVM-1 , the data for the bdpy alone was subtracted from the data obtained for IVM-bdpy ( Fig 8B ) . The time constants for the three plots were 6 . 5 min for the IVM-bdpy , 8 . 2 min for the bdpy fluorophore and 6 . 0 min for IVM-1 . As the lipophilic properties of IVM and IVM-1 are similar ( logPs of 5 . 4 and 5 . 1 , respectively ) , we infer this analysis quantitatively reflects the membrane partitioning properties of IVM . Notably , the resultant plot plateaued at about 18–20 minutes , which was similar to the estimated time to plateau for the IVM-activated current as presented in Fig 5A and 5B . It should be noted , however , that the IVM-bdpy experiment was done using a 100-fold higher concentration of drug in un-transfected cells and represents passive membrane accumulation , whereas the current plateau involved IVM binding to and activating receptors . Nevertheless , these data suggest that membrane partitioning of IVM is the rate limiting factor controlling the activation rate of GluClRs . The second experiment was aimed at determining the lateral membrane diffusion properties of IVM using the fluorescence recovery after photobleaching ( FRAP ) technique . After a 30-min equilibration time , 5 cells were monitored for fluorescence recovery within a bleached circular patch of membrane of 3 . 7 μm diameter ( Fig 8C ) . Any bleaching and recovery of the surrounding area of the cell was measured , using a similar circular patch and used to correct the recovery rate of the membrane delineated by the bleached area . We also noted that the recovery reached 98 ± 2% of control . The resultant plot , fitted to a standard exponential , is shown in Fig 8D . The time constant ( τ ) for recovery of 1 . 3 minutes was used to calculate the diffusion coefficient ( D ) , using the radius ( r ) of the bleached patch , according to D = r2/4τ . The calculated value of D was 1 . 1 x 10−2 μm2s−1 . Over the course of 1 s , the IVM-bdpy probe would traverse 0 . 2 μm along the membrane surface ( lateral diffusion rate of 0 . 2 μms−1 ) . This value is comparable to that calculated using a similar probe that was used to measure partitioning and lateral diffusion in muscle membranes of Ascaris suum [39] . We infer that this slow lateral diffusion rate coupled to membrane partitioning are the major contributors to local increases of IVM around its target binding sites and the slow quasi-irreversible actions of IVM at GluClRs .
Resistance to IVM in pest species is a growing problem worldwide . Previous research efforts have identified mutations and allelles of GluClRs associated with resistance to IVM , however , there is a lack of information regarding the mechanisms that underlie resistance . In this study , we investigated the synaptic and biophysical properties of different GluClR subtypes and detailed a possible mode of resistance in the pest species H . contortus . We found that β subunits of H . contortus can assemble both as homomers and as heteromeric combinations with the α subunit . Given that both subunits have common distribution patterns in motor neuron commissures and nerve cords [14] , it is possible that these two subunits combine to form heteromeric receptors , which are less sensitive to IVM but retain high sensitivity to neurotransmitter . Our data also support the idea that the subunit stoichiometry of heteromeric GluClRs is variable and that receptors with a greater proportion of β subunits are less sensitive to IVM . With a high EC50 ( 400 μM ) for glutamate , a low unitary conductance and brief active periods , homomeric β GluClRs are not likely to be efficacious post-synaptic receptors . Moreover , given that our single channel recordings reveal little homomeric β GluClR activity in patches expressing both α and β subunits , homomeric β GluClRs are likely to be scarce in cells that also express α subunits . Using heterocultures of neurons and transfected HEK cells , we demonstrate that α and αβ GluClRs mediate IPSCs in response to presynaptic glutamate release . HEK293-neuronal co-cultures have been shown to recapitulate synaptic current profiles in excitatory glutamatergic synapses expressing NMDA or AMPA receptors [48] suggesting that glutamate release in heterosynapses is likely to be physiological . To our knowledge , no published reports are available describing the properties of IPSCs mediated by native GluClRs in invertebrate neurons or muscle cells . Our preparation therefore represents a technological advance with the potential to improve our understanding of ( 1 ) the kinetics of IPSCs mediated by defined GluClR isoforms , ( 2 ) the effects of drugs on IPSCs mediated by defined GluClR isoforms , ( 3 ) the effect of posttranslational modifications ( e . g . , phosphorylation ) and resistance mutations on the formation and function of synapses , and ( 4 ) synaptogenesis and synaptic clustering mechanisms . We observed a substantial increase in IPSC peak amplitude and a general slowing of IPSC kinetics , particularly the decay times . IPSC experiments also revealed a tonic inhibitory component , which is a well-studied mode of inhibition in vertebrates that may also involve extrasynaptic pLGICs [49] . In the presence of IVM we demonstrate that whereas the tonic component of inhibition is similar in cells expressing α and αβ GluClRs , the phasic IPSC component is potentiated to a smaller degree in αβ GluClRs . This lower sensitivity is mostly due to the intrinsic activation properties of β-containing receptors . Hence , the net inhibitory signal mediated by αβ GluClRs in the presence of IVM would be less than that of α GluClRs , thereby underlying a possible mechanism of IVM resistance . Our single channel recordings showed that β-containing GluClRs had briefer active periods and a smaller conductance , both of which render receptors less sensitive to the potentiating effects of IVM when compared to α GluClRs [35] . Single receptor active durations also accord well with ensemble decay times as revealed in macropatches [35] and IPSCs ( this study ) . GluClRs with faster decay times are potentiated to a lesser extent by IVM in absolute terms and thus produce less inhibitory input to target cells . The decay times for the α GluClRs were 67 ms ( macropatch ) [35] and 40 ms ( IPSCs ) . For GluClRs that exhibited reduced IVM sensitivity the decay times were 11 ms ( macropatch ) for α ( G36’A ) [35] and 17 ms ( IPSCs ) for the αβ GluClRs . It is also noteworthy that vertebrate pLGICs , such as glycine and GABA-gated receptors that are much less sensitive to IVM also exhibit relatively brief single receptor active periods and faster ensemble decay times [43 , 50] . We hypothesised that a noteworthy component of the mode of action of IVM was its interactions with cell membranes . A fluorescently tagged IVM ( IVM-bdpy ) was used to measure the time-course of drug accumulation and lateral diffusion in membranes that was compared to the onset of current potentiation . Single channel currents recorded from small , ~2 μM diameter patches equilibrated over 1–2 minutes [35] . IVM-bdpy required about 18 minutes to saturate whole cell membranes , which was similar to the time taken for baseline currents to plateau ( 17 minutes ) in our IPSC recordings . In addition to the concentrating effects of membrane partitioning , a slow lateral diffusion rate ( 0 . 2 μms−1 ) of IVM would also contribute to the apparent association rates of IVM to binding sites at GluClRs [46] . The recovery after bleaching reached near control levels , but was not at 100% for each cell , suggesting the presence of membrane compartments of relatively immobile lipid rafts [51] . Our experiments suggest that the dominant pathway for IVM to reach its binding sites at GluClRs is by membrane partitioning and diffusion rather than inducing slow conformational rearrangements to the receptors after binding directly from the aqueous compartment . Finally , we sought to identify a new IVM analogue with a differential effect on α and αβ GluClRs as proof of principle that this pharmacophore may be useful in refining anthelminthic treatments by targeting particular GluClR isoforms . We identified key receptor-drug interactions that determine drug potency that include ( 1 ) the disaccharide group of IVM and ( 2 ) the significance of the C-5 configuration of IVM , which is a key determinant of differential potency between GluClR isoforms . In summary , we describe the functional properties of α homomeric and αβ heteromeric GluclRs using conventional whole cell recording , single channel analysis and heterosynaptic analysis . Our key findings support the general inference that the intrinsic activation properties of the inhibitory pLGICs is a critical determinant and predictor of the potentiating potency of IVM . Informed by our findings , drug design strategy may be directed towards increasing the active durations of single receptors and slowing decay times of IPSCs in synaptic isoforms of GluClRs .
cDNAs encoding the α ( avr-14b ) ( pcDNA 3 . 1+ ) or β ( pUNIV ) GluClR subunits of H . contortus were nuclear injected into oocytes ( NASCO , WI USA ) or transfected into HEK293AD cells ( CellBank Australia ) using a calcium phosphate-DNA co-precipitate method . Synapse formation between neurons and HEK293 cells was promoted by co-transfecting neuroligin 2A or 1B and the cDNA encoding CD4 surface antigen was also added to the transfection mixture so as to identify transfected cells . The oocytes and HEK cells were used for experiments 2–3 days after the introduction of the cDNAs . All experiments were done at room temperature ( 22 ± 1°C ) . Xenopus laevis oocytes were harvested by surgical incision . The oocytes were then defolliculated with 1 . 5 mg/ml collagenase for 2 hours . Free oocytes were rinsed with calcium-free OR-2 solution containing ( in mM ) 82 . 5 NaCl , 2 KCl , 1 MgCl2 , 5 HEPES , pH 7 . 4 and mature stage V or VI oocytes were isolated for experiments . The DNAs coding for α and β subunits were nuclear injected into the oocytes using a Nanoliter 2000 microinjector ( WPI Inc ) at a ratio of ( α:β ) , 1:0 , 0:1 , 1:1 and 1:50 . The total amount of injected DNA in all combinations was 400 ng ml−1 , including the 1:50 ( α:β ) combination ( 8 ng of DNA encoding the α and 392 ng of DNA encoding β subunits ) . Injected oocytes were incubated in ND96 storage solution ( in mM ) 96 NaCl , 2 KCl , 1 MgCl2 . 6H2O , 1 . 8 CaCl2 , 5 HEPES , 50 μg ml−1 gentamicin , 2 . 5 sodium pyruvate , 0 . 5 theophylline , pH 7 . 4 at 16°C for 2–3 days before experiment . For the two-electrode voltage clamp recordings , each oocyte was secured in a cell bath that was continually perfused with ND96 recording solution ( ND96 storage solution without pyruvate , theophylline and gentamicin ) . Glutamate , IVM and IVM analogues were diluted in ND96 recording solution and were applied to the oocyte via bath perfusion . The two microelectrodes contained 3 M KCl and had resistances of 0 . 2–2 MΩ . Recordings were done using Clampex 10 . 2 software ( Molecular Devices ) at a clamped voltage of −40 mV . Currents were low-pass filtered at 200 Hz , sampled at 2 kHz using a Gene Clamp 500B amplifier and digitised by a Digidata 1440A interface . The IVM ( IVM B1a ) concentration-response experiments were standardised by using an application protocol that consisted of applying increasing concentrations of IVM for , respectively , 3 min , 3 min , 3 min , 2 min , 1 min , 1 min and 0 . 5 min . The IVM-induced currents were normalised to a saturating glutamate concentration of 5 mM ( see example , Fig 2 ) . Single-channel currents were recorded from outside-out excised patches at a clamped potential of −70 mV . The single channel conductance was calculated by dividing the mean current by the net driving force at a clamped potential of ‐70 mV , after accounting for reversal and liquid junction potentials [35] . Excised patches were continuously perfused via a gravity-fed double-barrelled glass tube , which contained extracellular bath solution containing ( in mM ) , 140 NaCl , 5 KCl , 1 MgCl2 , 2 CaCl2 , 10 HEPES , and 10 D-glucose and titrated to pH 7 . 4 . Glutamate and IVM were dissolved in this extracellular solution . Recording electrodes were pulled from borosilicate glass ( G150F-3; Warner Instruments ) , coated with a silicone elastomer ( Sylgard-184; Dow Corning ) and heat-polished to a final tip resistance of 4–15 MΩ when filled with an intracellular solution containing ( in mM ) 145 CsCl , 2 MgCl2 , 2 CaCl2 , 10 HEPES , and 5 EGTA , pH 7 . 4 . Stock solutions of L-glutamate were also pH-adjusted to 7 . 4 with NaOH . A 10 mM stock of IVM ( Sigma-Aldrich ) was dissolved in 100% DMSO and kept frozen at ‐20°C . Fresh working stocks of IVM at 5 nM were prepared by dissolving the appropriate quantity directly in extracellular solution . 100% DMSO when dissolved in extracellular solution alone at the same concentration as was present in working solutions containing 5 nM IVM had no effect on patches excised from cells transfected with GluClRs or from untransfected cells . Currents were recorded using an Axopatch 200B amplifier ( Molecular Devices ) , filtered at 5 kHz and digitized at 20 kHz using Clampex ( pClamp 10 suite , Molecular Devices ) via a Digidata 1440A digitizer . Primary neuronal cultures were prepared from cortices of E18 rat embryos , ( University of Queensland Biological Services ) , which were triturated and plated at 100 , 000 cells per 18-mm poly-D-lysine-coated coverslip in DMEM with 10% fetal bovine serum . After 24 h the entire medium was replaced with Neurobasal medium that included 2% B27 and 1% GlutaMAX supplements . After one week half of this medium was replaced and the neurons were allowed to grow in vitro for 3–5 weeks before introducing the transfected HEK293 cells . Recordings of synaptic currents were done in whole-cell configuration at ‐70 mV using an Axopatch 200B amplifier ( Molecular Devices ) , filtered at 5 kHz and digitized at 20 kHz using Clampex ( pClamp 10 suite , Molecular Devices ) via a Digidata 1440A digitizer . Membrane partitioning and FRAP experiments were performed on an LSM 710 inverted 2-photon confocal microscope equipped with a 40X water immersion objective lens ( 1 . 2 NA/280 μM WD/0 . 208 μM/pixel and a GFP/Alexa 488 filter set . Untransfected HEK293 cells at a confluency of 50–75% were plated onto 35-mm glass bottom dishes one day prior to experiments . The IVM-bdpy fluorophore was dissolved in 100% DMSO and stored at ‐20°C at a concentration of 24 mM . A Mai Tai eHP 2-photon laser/760-1040 nm was used as an excitation source ( power at 920 nm ) for the FRAP experiments . The laser and bleaching power were set to 7% and 75% respectively . Membrane partitioning images were taken for 35 min at 0 . 5 Hz after replacing the cell medium ( extracellular solution ) with one containing the IVM-bdpy probe at a concentration of 500 nM . FRAP images were taken at a frequency of 0 . 5 Hz for 10 min after a 30 min equilibration period in 500 nM IVM-bdpy . Group data were analysed in SigmaPlot 13 . 0 using one-way ANOVAs , where p < 0 . 05 was taken as the significance threshold and expressed as mean ± SEM . Tests for normally distributed data are built into the SigmaPlot software . Confocal images were captured and analysed with Zen 2012 SP2 ( Zeiss ) software . Oocyte concentration-response data were fit to a Hill equation to obtain an EC50 and Hill co-efficient for each oocyte recording . These parameters were then averaged across multiple oocyte experiments of the same type . Single channel recordings were analysed in QuB software . Single channel currents were idealised and separated into discrete activations by applying critical shut times of 50 ms in glutamate alone or 50–100 ms in glutamate plus IVM to separate and define single receptor active periods . Patches yielded between 30–250 individual active periods . Critical shut times were determined by generating an initial shut duration histogram to continuous data that included inactive periods corresponding to receptor desensitisation . The selected critical times eliminated periods of receptor desensitisation while retaining singe receptor active periods . Mean active period duration and intra-activation open probability was estimated from each patch . Group means were obtained by averaging across multiple patches for each recording condition . | Glutamate-gated chloride channel receptors ( GluClRs ) mediate chemoelectric inhibition in invertebrate animals and are targets for broad-spectrum pesticides such as ivermectin . However , resistance to ivermectin threatens the effective control of invertebrates that cause a range of agricultural and human diseases . This study investigates different isoforms of GluClR expressed by the major agricultural endoparasite , Haemonchus contortus , on a synaptic and single receptor level . We discovered that ivermectin enhances synaptic current amplitude and decay and lengthens single receptor activity . Furthermore , ivermectin is less efficacious at GluClRs that incorporate a naturally ivermectin-resistant subunit , suggesting a potential resistance mechanism . Finally , we identify two chemical interactions between the GluClR and ivermectin that determine its potency and show that ivermectin binds to GluClRs via cell membrane interactions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"invertebrates",
"single",
"channel",
"recording",
"neurochemistry",
"caenorhabditis",
"vertebrates",
"social",
"sciences",
"neuroscience",
"animals",
"xenopus",
"germ",
"cells",
"animal",
"models",
"oocytes",
"caenorhabditis",
"elegans",
"model",
"organisms",
"amphibians",... | 2019 | GluClR-mediated inhibitory postsynaptic currents reveal targets for ivermectin and potential mechanisms of ivermectin resistance |
Cellular metabolism continuously processes an enormous range of external compounds into endogenous metabolites and is as such a key element in human physiology . The multifaceted physiological role of the metabolic network fulfilling the catalytic conversions can only be fully understood from a whole-body perspective where the causal interplay of the metabolic states of individual cells , the surrounding tissue and the whole organism are simultaneously considered . We here present an approach relying on dynamic flux balance analysis that allows the integration of metabolic networks at the cellular scale into standardized physiologically-based pharmacokinetic models at the whole-body level . To evaluate our approach we integrated a genome-scale network reconstruction of a human hepatocyte into the liver tissue of a physiologically-based pharmacokinetic model of a human adult . The resulting multiscale model was used to investigate hyperuricemia therapy , ammonia detoxification and paracetamol-induced toxication at a systems level . The specific models simultaneously integrate multiple layers of biological organization and offer mechanistic insights into pathology and medication . The approach presented may in future support a mechanistic understanding in diagnostics and drug development .
Human metabolism is an integral component of whole-body physiology and its dysfunction plays a key role in many systemic diseases . Frequent symptoms of metabolic diseases are changes in exometabolism [1] , [2] which usually follow upstream alterations in intracellular flux distributions [3] . In order to associate diagnostic observations at the organism level accompanying specific diseases with structural impairment at the cellular level , a mechanistic understanding of genotype-phenotype correlations is essential [3] , [4] . Adequate analytical methods for a systemic consideration of the underlying processes are still missing . However , such multiscale approaches are necessary to understand the highly complex and intertwined structure of biological networks and the interplay with the surrounding organism [3] , [5] , [6] . In recent years , modeling approaches have been developed describing biological processes at different levels of physiological organization based on multiple , divergent mathematical formalisms [5] , [7] , [8] , [9] . At the whole-body level , physiologically-based pharmacokinetic ( PBPK ) modeling quantitatively describes the absorption , distribution , metabolization and excretion ( ADME ) of endogenous and exogenous compounds within mammalian organisms [10] , [11] , [12] , [13] . In contrast to classical pharmacokinetic ( PK ) /pharmacodynamic ( PD ) modeling [14] , PBPK models aim for a mechanistic representation of ADME-related processes . Structurally , PBPK models consist of compartmental representations of all relevant tissues and the vascular system . Most notably , PBPK models are based on large amounts of prior anatomical and physiological information as well as generic distribution models , such that most model parameters can be either obtained from database collections integrated in the modeling software or they can be deduced from the physicochemistry of the compound [15] , [16] , [17] , [18] , [19] . Hence , even though PBPK models contain more than hundred ordinary differential equations and several hundred variables , the number of independent parameters which need to be adjusted during model development is small ( usually less than 10 , see also Materials and Methods ) . ADME-related processes can automatically be quantified based on compound-deduced parameters allowing a detailed representation of mass transfer across various tissue compartments . PBPK models have previously been used for mechanistic analyses of drug pharmacokinetics [20] , pharmacogenomics [21] , species extrapolation [22] or analysis of rare adverse events [23] . For analyses at the cellular level , metabolic network reconstructions are an important tool of bottom-up systems biology . Cellular metabolism gathers a multitude of upstream regulatory events onto the various layers of cellular organization such as the transcriptome and metabolome representing an important angle point in the physiology of an organism . Metabolic networks are typically described by stoichiometric matrices and intracellular flux distributions are inherent variables in such models . First stoichiometric models on human metabolism at genome-scale encompassed generic collections of metabolic biochemistry in human cells [24] , [25] . Recent models explicitly account for network structure in specific tissues thereby enabling , for the first time , the consideration of metabolic models within a specific context of human physiology [4] , [26] . While metabolic network models are applicable to the investigation of in vitro experiments with more or less well-defined media conditions , they do not suffice for considerations of in vivo metabolism , where the cell is embedded in the ever-changing environment of the surrounding tissue and organism . Therefore , human metabolism can only be fully understood by an integrative analysis which simultaneously considers the whole-body context . This allows in particular the quantification of cellular boundary conditions and the interference with intracellular states and processes . Several approaches for combining models covering different levels of biological organization have been described before [8] , [27] , [28] . With regard to metabolic networks , dynamic flux balance analysis ( dFBA ) has been used to couple stoichiometric models of metabolism with dynamic models of microbial batch cultures and integrated omics networks [29] , [30] , [31] . We here apply dFBA to describe human metabolic networks within the context of whole-body PBPK models ( Figure S1 in Text S1 ) . The approach allows the representation of human metabolism under simultaneous consideration of quantitative availability of substances at the organism level ( Figure 1 A ) . We exemplarily use HepatoNet1 [26] , a genome-scale model of human hepatic metabolism to analyze specific responses of the network in the face of time-dependent concentration profiles in liver tissue . Following this approach we investigate three application examples ( Figure 1 B ) . First , we use a multiscale PK/PD model to analyze the distribution and therapeutic effect of allopurinol in the treatment of hyperuricemia . In a second example , we consider the effect of impaired ammonia metabolism on blood plasma levels to demonstrate the methods' capability to identify biomarkers specific for pathologic changes in the metabolic state [25] , [32] , [33] . Finally , we apply our approach to the analysis of paracetamol-induced toxication on liver function .
PBPK models describe the processes underlying the distribution of a compound within the body based on prior physiological information and generic distribution models . Organs in PBPK models are usually subdivided in further compartments such as the vascular , interstitial and intracellular space [12] , [34] . The basic differential equations within these compartments describe uptake , secretion , formation and consumption of a particular compound , representing overall mass balance equations [10] . In contrast , stoichiometric models describe mechanisms within the cell at a much finer spatial scale , providing a more detailed insight in intracellular processes with a particular focus on cellular biochemistry . Thus , the intuitive point of contact between both model formalisms is the intracellular space , where PBPK models quantitatively describe time-concentration profiles of endogenous or exogenous compounds , which in turn represent substrates or products of metabolic networks at the molecular level . In order to relate the distribution of endogenous and exogenous compounds at the organism level to metabolic network structures and thus to a specific enzymatic process at the cellular level , the stoichiometric network was embedded in the dynamic whole-body model by step-wise model discretization . To this end , functional adaptation of metabolism , ultimately quantified by intracellular flux distributions and extracellular exchange rates , can be assumed to be fast in relation to the surrounding distribution processes at the whole body scale . Consequently , flux distributions are kept constant over each time interval [29] , [30] , [31] . In our case , the chosen time interval was 1 step/min . Hence , for a specific distribution of extracellular concentrations at a given point in time , intracellular steady state ( i . e . equilibrium ) can be assumed , and flux balance analysis ( FBA ) can be applied for the estimation of flux distributions [31] . Following the rational of network validation as used in HepatoNet1 [26] we here applied case-specific objectives such as maximization of ammonia production or maximization of uric acid production to quantify extracellular exchange rates with regard to a specific set of boundary conditions . Notably , intracellular flux distributions of biological relevance can hardly be identified using these functional objectives since they rather evaluate the macroscopic behavior of the cell . In contrast , the underlying flux space is assessed qualitatively . In our approach , a compound in the PBPK model can act either as a regulatory modifier or as a substrate of an enzymatic reaction in the metabolic network . We therefore considered two distinct ways of coupling PBPK models and stoichiometric network models: ( 1 ) indirect coupling , where concentrations of a compound in the PBPK model impose a regulatory effect on enzyme activity which is quantified at the cellular level ( ‘feed-forward’ ) , thereby restricting fluxes through this specific reaction and ( 2 ) direct coupling , where perturbed metabolic processes ( for instance inhibited enzymes ) iteratively affect availability of a substance in the PBPK model by directly interfering the corresponding mass balance ( ‘feed-back’ ) . In both cases , the intracellular concentration of a compound constrains a metabolic state in the underlying network structure [35] . This also influences further downstream events , since the catabolic or anabolic products formed within the intracellular metabolic network are again distributed at the whole organism level . This centralized consideration of metabolism as a core component in human physiology can be seen as an hourglass or bow-tie scheme ( Figure 1 A ) [36] , [37] . In particular , enzymatic blockage results in accumulation of the upstream substrate , depletion of the downstream product and potential activation of alternative pathways . Details for indirect and direct coupling will be explained in the following . As a first example we analyzed drug action of allopurinol in the treatment of hyperuricemia with a multiscale PK/PD model . Purine metabolism provides a large number of drug targets [38] , [39] with uric acid being the final downstream degradation product in the human body . Quantitatively modeling the effect of drugs affecting this crucial metabolic pathway therefore provides valuable insights for drug development . In clinical practice , high plasma levels of uric acid ( above 470 µM ) are referred to as hyperuricemia which may result from inborn errors of purine metabolism and even more often from impaired renal excretion of uric acid , which is considered in the following . Hyperuricemia can lead to diseases like gout where uric acid is deposited into tissues , especially in the joints [40] . A currently used drug against hyperuricemia is the purine analog allopurinol [41] , which reduces production of uric acid by inhibition of xanthine oxidase . This enzyme oxidizes hypoxanthine to xanthine and subsequently xanthine to uric acid [42] . Allopurinol itself is oxidized by xanthine oxidase to oxypurinol , which also inhibits xanthine oxidase as an active metabolite ( Figure S2 in Text S1 ) . Since metabolization of allopurinol to oxypurinol is very fast , but excretion of oxypurinol is very slow , oxypurinol plays a significant role in the inhibition of xanthine oxidase [41] . To estimate the inhibitory concentration of allopurinol and its active metabolite oxypurinol , a coupled PBPK model was developed ( see Materials and Methods ) . Experimentally measured plasma concentrations following oral administration of 200 mg allopurinol [43] were considered for identification and fine-tuning of four physiological parameters describing the absorption and clearance of both compounds as well as four physicochemical parameters of allopurinol and oxypurinol ( Section parameter identification in Text S1 , Table S1 in Text S1 ) . Notably , these parameters are distributed in the two independent PBPK models which are coupled by the clearance reaction . Exemplarily , a sensitivity analysis was performed to estimate the influence of each parameter to the model ( Figure S3 in Text S1 ) . The resulting coupled model simultaneously described the PK of allopurinol and oxypurinol with excellent accuracy ( Figure 3 A ) . Having established a working model for single dosing of allopurinol , the PK of multiple administrations was predicted in the next step ( Figure 3 B ) . When simulating concentration profiles of allopurinol and oxypurinol over a time range of 35 days it became obvious that allopurinol is not accumulating in the human body but is always completely degraded before the next application is given . In contrast , oxypurinol can no longer be completely removed from the body leading to mean oxypurinol concentrations of about 45 µM . In the next step , a PBPK model for uric acid was created , which allowed describing changes in uric acid concentration as a mechanistic consequence of allopurinol inhibition in hepatic purine metabolism ( Table S2 in Text S1 ) . The plasma concentration of uric acid in healthy male individuals is around 302 µM ( male: 302±60 µM , female: 234±52 µM ) , while patients with gout show much higher concentrations of approximately 480 µM [44] . This physiological information was used to identify three parameters for the steady state clearance and production rate of uric acid in a whole-body model of a healthy as well as a gouty male individual ( Table S2 in Text S1 ) . An identical uric acid production rate was assumed for both healthy and gouty individuals , since impaired renal excretion is assumed to be the physiological cause for hyperuricemia in the present case . To finally couple the upstream distribution of allopurinol and oxypurinol at the whole-body scale with the subsequent inhibitory effect on xanthine oxidase in the hepatic metabolic network , indirect coupling was used to simulate the PD effect of a single as well as multiple allopurinol doses ( Figure 3 C , D ) . Maximization of uric acid production was considered as objective function , and the resulting FBA problem was additionally constrained by drug-induced enzymatic inhibition as well as the uric acid formation rates estimated in the PBPK model . IC50 values of allopurinol ( 13 . 4 µM [45] ) and oxypurinol ( 15 . 6 µM [46] ) were used to estimate the Ki constants of the enzyme inhibition by using the Cheng-Prusoff equation ( Equations S1–S6 in Text S1 ) [47] . Taken together , the overall multiscale model comprised the two coupled whole-body PBPK models of allopurinol and oxypurinol , hepatic metabolism and the downstream whole-body PBPK model of uric acid . After successful establishment of the multiscale model , the development of the uric acid level in gouty patients monitoring the therapeutic success of gout treatment after single and multiple dosing of allopurinol was simulated ( Figure 3 E , F ) . Before allopurinol treatment , the mean gouty patient in hyperuricemic steady state had a venous plasma concentration of 476 µM . After a single application of allopurinol , uric acid concentrations began to decrease . Therapy interruption or patient non-compliance , however , led to a recovery of uric acid concentrations which are typical for the hyperuricemic state ( Figure 3 E ) . Only when multiple dosings of allopurinol were routinely taken , a continuous decrease in uric acid concentrations could be achieved with venous blood concentrations of 146 µM . Most notably , the predicted uric acid concentration after multiple dosings was close to the range of uric acid observed in vivo ( Figure 3 F ) [48] , [49] although the PBPK model was established based on a single dosing of allopurinol . This correct prediction of our model clearly emphasizes the predictive capabilities of the coupled model . As a second example we analyzed pathogenesis of urea cycle disorders at the cellular scale and the resulting effect on ammonia plasma concentration at the organism level . In particular we aimed for an evaluation of diagnostic markers for a specific health state or disease progression . Hepatic metabolization of amino acids and detoxification of ammonia play an important role in the human body . Up to 95% of ammonia metabolized in hepatocytes is degraded to urea , which is subsequently excreted , while about 5% are metabolized to glutamine and about 1% to alanine [26] . Impairment in ammonia metabolism leads to decreased ammonia elimination and thereby induces hyperammonemia [50] . A consequence of hyperammonemia is an increase of ammonia concentration in the brain - so-called hepatic encephalopathy - which can cause confusion , lethargy , disorientation and in severe cases coma and death [51] , [52] . Liver dysfunction in ammonia metabolism can be caused by liver diseases or inborn errors of metabolism ( IEMs ) , e . g . urea cycle disorders ( UCDs ) , which may have lethal consequences without adequate treatment and diet [53] . Perturbations in ammonia detoxification capacity causes direct downstream changes in blood metabolite concentrations making ammonia detoxification a primary example for the identification of disease specific biomarkers . In humans , ammonia is produced by the breakdown of amino acids in the liver or intense muscle exercise [54] . In addition to endogenous ammonia , exogenous ammonia also enters the body with nutritional intake . Altogether , about 17 g of ammonia are produced by the body every day . The excretion through the kidneys is about 13 g per day , while 4 g per day are metabolized by the liver . Furthermore , during impaired ammonia detoxification following UCDs , the rates of glutamine and alanine synthesis are increased 4–6 fold [53] , [55] . As a first step to investigate impairment of ammonia detoxification using the multiscale coupling approach , a PBPK model of ammonia was established ( Table S3 in Text S1 ) . To determine rates of ammonia formation and consumption , an equilibrium concentration of 29 µM in venous plasma was considered within the PBPK model , which is the normal level in healthy humans [48] . Three model parameters describing ammonia production and excretion were identified using above physiological information: An overall ammonia production rate of 0 . 694 µmol/L/min was estimated as well as macroscopic liver and kidney clearance rates of 0 . 163 µmol/L/min and 0 . 530 µmol/L/min , respectively . Next , the ammonia PBPK model and the metabolic network were combined by using direct coupling . In particular , the PBPK model was simulated for one time step to calculate the new concentrations and the corresponding liver clearance rate , which could then be used as a new upper bound in the next FBA step . Maximization of ammonia production was used as objective function , which was constrained by substrate availability , exchange rates calculated with the PBPK model as well as enzymatic deficiencies accompanying UCD . At a steady state ammonia concentration of 29 . 02 µM in the venous blood , the liver cell showed an intracellular ammonia concentration of 25 . 78 µM ( Figure 4 A ) . In the metabolic network of the hepatocyte , an ammonia uptake flux of 0 . 163 µmol/L/min was calculated , while the production rates of urea , glutamine and alanine were 0 . 070 µmol/L/min , 0 . 008 µmol/L/min and 0 . 002 µmol/L/min , respectively ( Figure 4 B ) . The demands of glucose and oxygen are in agreement with previous results [26] . With the combined multiscale model , we simulated pathogenesis of urea cycle disorder ( UCD ) resulting in a reduced ammonia detoxification capacity . Pathogenesis of UCD was exemplarily assumed as a linear decrease of the enzyme activity of ornithine transcarbamylase ( OTC ) , leading to complete impairment of the enzyme ( Figure 4 A ) [56] . Simultaneously , the glutamine and alanine production rates are increased fourfold above the nominal glutamine and alanine production rates [57] . Hence , while the enzyme activity of OTC is decreased , the maximum activity of glutamine and alanine synthesizing enzymes is increased . Glutamine and alanine production is supposed to increase at 6 h after the onset of UCD , assuming a delay due to transcription and translation initiation . With the coupled model and the above described constraints , simulation was performed within a time range of 21 days . After 24 h , UCD starts developing which led to a decreasing urea production rate after 30 h . At the same time , the glutamine and alanine fluxes began to rise and the rate of ammonia uptake began to decrease . In the new steady state ( after 6 . 5 days ) , ammonia uptake rate was 0 . 072 µmol/L/min ( Figure 4 B ) and glutamine and alanine production rates were 0 . 033 µmol/L/min and 0 . 007 µmol/L/min , respectively ( Figure 4 C ) . The new venous ammonia concentration was 33 . 99 µM , while the liver concentration was 30 . 27 µM ( Figure 4 A ) . Above model is a representation for an average patient . Since the model , however , structurally includes many potential causes for inter-individual variability at a large level of mechanistic detail , it may be used to analyze pathogenesis on a populations scale as well . Such individual differences may include physiology [58] , protein expression [20] or even nutrition [59] . In order to exemplarily describe the effect of inter-individual variability , 100 individuals were simulated based on randomly perturbed production and clearance rates , respectively ( assuming 10% standard normal distribution relative to the mean patient , Figure 5 A , Table S4 in Text S1 ) . The distribution of ammonia concentrations in healthy and diseased individuals together with their cumulative sums underlines the inter-individual variability during UCD pathogenesis ( Figure 5 B ) . Performing the Kolmogorov-Smirnov test provided evidence that the distributions in healthy and diseased state differs significantly ( p<0 . 001 ) , making ammonia concentration a quantitative biomarker for OTC deficiency . By performing the population simulation , the results demonstrate that the mere consideration of single individuals may induce misleading diagnoses when specific patient subgroups such as obese , elderly or diseased individuals are to be investigated . Only by population simulations of comprehensive mechanistic models it may become possible to mechanistically discriminate the different , potentially counter-current factors such as high ammonia production rates and low clearance rates . Our final example deals with drug-induced toxication by addressing downstream effects of drug dosing on metabolic functionality at the cellular level . We here chose overdosing of paracetamol , one of the most common reasons for liver failure [60] and poisoning [61] . Paracetamol is generally considered to be an inhibitor of prostaglandin synthesis and is widely used for reducing pain ( analgesic ) and fever ( antipyretic ) [62] . At higher doses , paracetamol can cause severe hepatotoxic effects leading to acute liver failure and liver necrosis [63] . Paracetamol is metabolized by three main pathways: ( i ) glucuronidation , ( ii ) sulfation and ( iii ) N-hydroxylation by cytochrome P450 2E1 ( CYP2E1 ) [64] . The corresponding metabolites of the three pathways are ( i ) paracetamol glucuronide ( PG ) , ( ii ) paracetamol sulfate ( PS ) and ( iii ) N-acetyl-p-benzoquinone imine ( NAPQI ) . In particular , NAPQI , the hydroxylation product , is relevant for consideration of paracetamol toxication since it is highly reactive and toxic [63] . At therapeutic doses , NAPQI is almost immediately detoxified by glutathione ( GSH ) conjugation . After reacting with GSH , NAPQI is further degraded in the gut and the kidneys and is excreted as paracetamol cysteine ( AC ) and mercapturic acids [65] . Paracetamol is metabolized primary into PG and PS . At higher doses , however , the pathways synthesizing PG and PS become saturated causing more NAPQI to be produced [64] . In this case , GSH is depleted almost completely ( up to 80% ) by the detoxification of NAPQI such that the excess of NAPQI accumulates in the liver and the body . The free NAPQI then binds covalently to proteins forming protein adducts which are considered to be one cause for hepatotoxicity of paracetamol [63] . Taken together , the metabolic impact of paracetamol overdose is described by the reduced activity of three enzymes N-10-tetrahydrofolate dehydrogenase ( THFDH , up to 25% ) , glutamate dehydrogenase ( GDH , up to 25% ) and mitochondrial ATP-Synthetase ( ATPS , up to 60% ) , respectively , and the depletion of GSH by up to 80% [66] . To quantify the metabolic impact of perturbations in the enzyme activities on whole liver functionality , indirect coupling of a paracetamol PBPK model and HepatoNet1 was used to determine the effect of a paracetamol overdose on a large amount of functional liver objectives . HepatoNet1 was previously validated with 123 physiological functions which represent essential tasks for liver metabolism [26] . In particular , 67 of the 123 presented objectives have been tested with three specific sets of extracellular metabolites , which were therefore used as a core set for network validation [26] . To quantify the impact of paracetamol-induced liver failure , we tested the extent by which the value of each of the 67 objective functions is decreased during a paracetamol overdose , thus quantifying hepatic network robustness towards external perturbations [67] . We started our analyses with the development of PBPK models of the parent drug paracetamol the three metabolites ( Table S5 in Text S1 ) . Oral administration of 1 g paracetamol was considered first . Eight model parameters were identified for the paracetamol model , three for each the PG and the PS model and four for the NAPQI model by comparison of computational simulations with the corresponding experimentally measured venous blood concentrations for all four compounds [65] ( Section parameter identification in Text S1 , Table S5 in Text S1 ) . Notably , the parameters either characterize physicochemistry of the compounds or describe the physiology of the individuals such that prior knowledge is implicitly included ( see Materials and Methods ) . Moreover the four models are highly interlinked by the underlying mass-balances thereby reducing the systemic degree of freedom significantly . Since the metabolization of NAPQI into AC cannot be quantified due to missing literature information , it was assumed that the AC concentration is equivalent to the NAPQI concentration . After parameter adjustment , the PBPK simulations of paracetamol , PG , PS and NAPQI described the experimental data with excellent agreement ( Figure 6 A , Table S6 in Text S1 ) . Subsequently , the PK of the four compounds after a lethal dose of 15 g paracetamol was predicted ( Figure 6 B , Table S7 in Text S1 ) . In order to investigate changes in metabolic functionality following a paracetamol overdose , FBA was first performed to determine the optimal value of every objective function and to quantify all fluxes in the healthy , untreated reference individual . Subsequently , the inhibition of enzymes by paracetamol and its active metabolites was implemented as additional constraints on the flux values . For every inhibited enzyme the flux value was fixed as the product of the value in the healthy state and the remaining relative enzyme activity ( Equation 6 ) . The Ki values were calculated by assuming that the maximum concentration of NAPQI after a lethal dose of 15 g paracetamol induces the maximum enzyme inhibition as described above . In order to calculate Ki values for paracetamol and all three metabolites , substrate concentrations were assumed to be equal to Km ( Equations S7–S12 in Text S1 ) . Therefore , the time-resolved relative enzyme activities ( Figure 6 C , D ) could be calculated by: ( 6 ) FBA was then performed for every objective function with these additional constraints . Since NAPQI is not naturally produced in the liver , the molecular interaction of exogenous NAPQI with endogenous GSH cannot be implemented in HepatoNet1 without structural network modifications . Hence , the production of NAPQI as an exogenous compound is simulated in the PBPK model as described above . The depletion of GSH was considered rather phenomenological by indirect coupling , thereby inducing a decrease in enzyme activity which is linked by indirect coupling to the intracellular NAPQI concentration . Neither metabolite pool size of GSH nor regulatory effects following depletion of GSH can be mechanistically considered in stoichiometric models such as HepatoNet1 . Therefore , all reactions producing GSH ( Table S8 in Text S1 ) in HepatoNet1 were identified and inhibited as described above , such that GSH consuming reactions are limited . The change in liver functionality was calculated as the difference between the maximum values of the objective functions in the healthy state and in the case of a paracetamol overdose . Three distinct time points were considered which included time of peak concentrations of paracetamol ( tmax , paracetamol ) and NAPQI ( tmax , NAPQI ) , respectively as well as time of trough concentrations at 24 h after drug administration ( t24h ) ( Figure 6 E , F ) . Out of the 67 objective functions considered , 20 objective functions underwent a change after the application of 1 g paracetamol with respect to the untreated state , while 24 objective functions underwent a change after the application of 15 g paracetamol . Almost all optimal values were at reference values after 24 h for a 1 g dose of paracetamol , while all affected objectives remained severely decreased after 24 h for a 15 g dose . We next analyzed , whether the flux underlying the different optimal values changed during the application of different doses of paracetamol . This could explain why more objective functions are affected after the toxic dose of paracetamol , which is to be expected by network robustness [67] . As an example of an objective function which is only affected after the toxic dose , the production of oleate showed a greater robustness to the metabolic effect of paracetamol , as many active fluxes underlying this objective function remained unchanged after the application of 1 g paracetamol ( 54 . 8% ) ( Figure 7 A ) . Likewise , the maximum value of the objective was only slightly decreased after the application of 15 g paracetamol . In contrast , paracetamol administration showed a greater effect to the maximum value of cysteine production , as the maximum value was already decreased after an application of 1 g paracetamol compared to the untreated reference state . Furthermore , only 11 . 1% of the fluxes underlying this objective function remained unchanged after paracetamol administration and many new fluxes became active , suggesting that flux rerouting [68] was used to compensate the inhibition in the impaired metabolic pathway ( Figure 7 B ) .
We here integrate genome-scale metabolic networks , comprising thousands of biochemical reactions , into PBPK models to analyze metabolism at the level of the human organism . The approach is based on dFBA [31] which is used to simulate stationary stoichiometric metabolic networks at the cellular level in combination with ODE-based PBPK models . This semi-continuous approach uses the metabolic exchange rates as input at the organism level and vice versa the drug concentrations at the whole-body scale to calculate upper bounds for the enzyme activities in the metabolic network . Our approach combines well-established computational modeling approaches from different biological scales . At the organism scale , standardized PBPK models which are routinely used in pharmaceutical drug development provide a generalized description of the distribution of substances within the human body . At the cellular scale , metabolic network reconstructions represent core building blocks of bottom-up systems biology which describe fundamental cellular biochemistry . Our coupling approach therefore provides a generic framework for a wide range of possible applications . Notably , all these applications can be addressed without the need for further model curation or modification . To illustrate our approach , we exemplarily integrate HepatoNet1 [26] , a genome-scale metabolic model of a human hepatocyte into the liver tissue of a standardized PBPK model [11] , [69] . As in reality , metabolization processes at the organism level therefore result from the biochemical reaction within the hepatocyte . To illustrate the broad applicability of our approach , three case studies are presented addressing prototypical medical and pharmaceutical questions . In a first application example we investigate a multiscale PK/PD approach , where indirect coupling is used to mechanistically describe the pharmacological effect of the purine analogue allopurinol on the biosynthesis of uric acid . The PK of the exogenous drug as well as its resulting downstream PD effect on the formation of endogenous metabolites are quantified allowing a comprehensive evaluation of drug safety and drug efficiency . For multiple oral administrations of 200 mg allopurinol we predict a 69 . 3% decrease of uric acid concentration in the venous plasma which is in quantitative agreement with clinical data [48] . Since the corresponding PK/PD model has been established with data from single dosings of allopurinol [43] , this accurate prediction of the long term therapeutic effect convincingly illustrates the predictive power of our approach . The identification of quantitative biomarkers for metabolic disorders is the second application example . As a complement to classical qualitative biomarker identification [70] , our approach provides quantitative information in terms of specific concentration profiles by simulating the distribution of the affected compounds at the organism level . This enables a mechanistic description of metabolic disorders such as IEMs in blood plasma and further biofluids . As an exemplary case study for quantitative biomarker identification , we investigate impaired ammonia detoxification resulting in an increase of 17% of ammonia in blood plasma . We next simulate virtual populations of healthy individuals and patients by varying anatomical and physiological parameters according to prior statistical information [58] . Despite a considerable level of inter-individual variability in both subgroups , we demonstrate that the difference in ammonia plasma levels between both subgroups is statistically significant . Notably , the structural complexity of PBPK models together with the prior physiological and anatomical information included helps to explain counter-intuitive behavior during disease progression in individual patients , since many relevant co-factors are mechanistically presented in the model structure itself . This is important in clinical practice , since in addition to diagnostics of key metabolites other contributing factors are structurally considered in the model such that variability of the healthy reference state can be mechanistically quantified . Drug-induced toxication following paracetamol overdose is the third application example . We demonstrate how paracetamol impairs metabolic capacity by affecting a broad range of different metabolic functions . The specific metabolic impact is illustrated for a therapeutic ( 1 g ) and a toxic ( 15 g ) dose of paracetamol . The results show a larger metabolic impact on single hepatic functions following the toxic dose of paracetamol . Also , more metabolic functions are affected after the higher dose ( 24 vs . 20 functions ) . Since a considerable number of fluxes is affected , we conclude that paracetamol toxication induces flux rerouting which is used by the hepatocyte to compensate for network perturbations thereby conferring cellular robustness . Notably , this effect which has been shown before for microorganisms [68] is distributed over the whole metabolic network and can only be investigated with genome-scale models . Describing the impact of a compound on the body and the cell is the fundamental question in pharmacodynamics . Classical approaches describe this interference with rather phenomenological models [14] . By replacing the cellular space with metabolic network models at genome-scale , our approach describes cellular processes at a much higher level of detail . To quantify specific metabolic states we use objective functions which have been used before to verify functional capacity of the liver [26] . It should be noted that such functions do not allow to identify actual intracellular flux distributions unlike shown before for microorganisms [71] . In contrast , the flux space is evaluated qualitatively in the face of external or internal perturbations . Additionally constraining FBA optimization with kinetic rates simulated with the PBPK models , however , provides an important transfer of physiological information in-between both model scales . If it would be possible to quantify intracellular flux distributions in mammalian cells as well , further algorithms [72] could be used to further asses and characterize the overall flux space . This may also involve the consideration of additional experimental information such as omics data [73] or inclusion of regulatory information [74] . Taken together , the presented approach integrating organ-specific metabolic networks into PBPK models provides many opportunities for scientific research , clinical applications and drug development as outlined by the prototypical examples discussed above . It is only by such multiscale models that a mechanistic understanding of organ dysfunction and disease etiology at a system level will be achieved . This will be greatly supported by the reconstruction of further genome-scale metabolic networks which will become available in the future [7] . Structurally , the metabolic networks provide a template onto which the genetic predisposition of a patient can be mapped . Together with specific physiological information this may someday allow model-based optimizations of risk-benefit profiles in personalized medicine .
At the whole-body level , PBPK modeling quantitatively describes all ADME-related processes of endogenous and exogenous compounds within mammalian organisms [10] , [11] , [12] , [13] . In contrast to the rather descriptive consideration of PK in classical compartmental models , PBPK models include a detailed representation of physiological processes within an organism which is based on prior knowledge and information . The underlying model structure , which connects the various tissues compartments and the vascular system , is based on generic distribution models and quantifies the mass transfer across the different sub-compartments . Parameters of distribution models are automatically derived from the physicochemistry of the compounds such as the molecular weight or the lipophilicity [15] , [16] , [17] , [18] , [19] . Parameters describing the physiology of an organism such as organ volumes , blood flow rates or tissue composition are obtained from collections of physiological data and are available in the internal PBPK software database . Due to this large amount of prior structural and physiological information , the number of independent parameters which need to be identified during model development is small ( usually less than 10 ) . Notably , there is a clear separation between physiological parameters which refer to the organism in which a compound is distributed and compound specific parameters which specifically describe the properties of the substance itself . Thereby , compound parameters only need to beadjusted in a narrow range since literature information already is available . In contrast , physiological parameters have to be identified individually . The PBPK models considered in this study were all built with the software tools PK-Sim and MoBi for which academic licenses are available free of charge ( Section software information in Text S1 ) and which have been explained in detail before [11] , [69] . PBPK models were created for parent substances as well as for all metabolites . Compound specific parameters were used in each case to parameterize the underlying structure of the PBPK model . In all models , we considered mean individuals . The anthropometric information regarding age , weight and height further specifies the selection of physiological parameters as provided in the software . This allows a specific parameterization of the PBPK model , since the model contains only few independent parameters as described above . Case-specific metabolization and clearance processes represented by Michaelis-Menten kinetics or first order reactions and according to literature information are additionally defined . For the corresponding kinetic parameters , parameter adjustment is important since ( 1 ) usually less literature data is available and ( 2 ) these parameters often have significant sensitivity although the overall dynamic behavior of the PBPK models is robust ( Figure S3 in Text S1 ) . For the examples of allopurinol treatment and paracetamol toxication , PBPK models in PK-Sim are exported to MoBi where they are further modified . PBPK models of the parent drug and its metabolites are connected such that the distribution of all compounds is described simultaneously . In this case , the rate of the clearance reaction in the parent drug model is set as the input into the metabolite model . Stoichiometric models provide a mathematically formal way to capture the basic biochemistry of cellular metabolism into an analytical framework . Assuming steady state of the system , all intracellular metabolites can be balanced in linear systems of equations , which are usually underdetermined since they encompass much more unknown reaction rates than linear independent mass balance equations . Flux distributions can be identified with stoichiometric models by using FBA [35] , [75] , [76]: ( 4 ) Here , f · v ( , ) corresponds to the objective function which reflects a rationale of cellular function , S ( ) denotes the stoichiometric matrix of the metabolic network ( including m metabolites and n reactions ) and v represents the intracellular flux distribution . The overall solution space is confined by a set of additional constraints ( A·v≤b ) which represent for example substrate availability . To illustrate our approach , we here exemplarily use HepatoNet1 , a tissue-specific network reconstruction of hepatic metabolism [26] . HepatoNet1 consists of 777 metabolites and 2539 reactions and is divided into six intra- and two extracellular compartments . Its general network structure was validated by verifying the accomplishment of 123 biochemical objectives representing metabolically feasible functional modes . Hence , the network was constructed specifically to examine distinct metabolic processes of the liver and provides a structural platform for mechanistic studies of tissue-specific physiological functions [26] . | Cellular metabolism is a key element in human physiology . Ideally the metabolic network needs to be considered within the context of the surrounding tissue and organism since the various levels of biological organization are mutually influencing each other . To mechanistically describe the interplay between intracellular space and extracellular environment , we here integrate the genome-scale metabolic network model HepatoNet1 at the cellular scale into physiologically-based pharmacokinetic models at the whole-body level . The resulting multiscale model allows the quantitative description of metabolic behavior in the context of time-resolved metabolite concentration profiles in the body and the surrounding liver tissue . The model has been applied to three case studies covering fundamental aspects of medicine and pharmacology: drug administration , biomarker identification and drug-induced toxication . Most notably , our multiscale approach fosters an improved quantitative understanding of drug action and the impact of metabolic disorders at an organism level , based on a genome-scale representation of cellular metabolism . Computational models such as the one presented include various aspects of human physiology and may therefore significantly support rational approaches in medical diagnostics and pharmaceutical drug development in the future . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"drug",
"absorption",
"drugs",
"and",
"devices",
"metabolic",
"networks",
"population",
"modeling",
"pharmacology",
"drug",
"metabolism",
"drug",
"distribution",
"biology",
"metabolic",
"disorders",
"pharmacokinetics",
"systems",
"biology",
"computational",
"b... | 2012 | Integrating Cellular Metabolism into a Multiscale Whole-Body Model |
Human kidney function declines with age , accompanied by stereotyped changes in gene expression and histopathology , but the mechanisms underlying these changes are largely unknown . To identify potential regulators of kidney aging , we compared age-associated transcriptional changes in the human kidney with genome-wide maps of transcription factor occupancy from ChIP-seq datasets in human cells . The strongest candidates were the inflammation-associated transcription factors NFκB , STAT1 and STAT3 , the activities of which increase with age in epithelial compartments of the renal cortex . Stimulation of renal tubular epithelial cells with the inflammatory cytokines IL-6 ( a STAT3 activator ) , IFNγ ( a STAT1 activator ) , or TNFα ( an NFκB activator ) recapitulated age-associated gene expression changes . We show that common DNA variants in RELA and NFKB1 , the two genes encoding subunits of the NFκB transcription factor , associate with kidney function and chronic kidney disease in gene association studies , providing the first evidence that genetic variation in NFκB contributes to renal aging phenotypes . Our results suggest that NFκB , STAT1 and STAT3 underlie transcriptional changes and chronic inflammation in the aging human kidney .
The human kidney shows a steady and quantifiable decline in function with age , accompanied by stereotyped changes in gene expression and histopathology [1] . The glomerular filtration rate ( GFR ) , a clinical measure of kidney function , shows a steady decline in most individuals starting at about age 40 [2 , 3] . Kidney aging is an important public health problem because the age-related decline in GFR can lead to chronic kidney disease and progression to end-stage renal disease [3 , 4] . Chronic kidney disease is a common age-related condition; 35–40% of people over the age of 70 in the United States have some form of chronic kidney disease [3] . Moreover , reduced kidney function is associated with an increased risk of cardiovascular disease [5] . Kidney aging is accompanied by several characteristic changes in renal histopathology including tubulointerstitial fibrosis ( extracellular matrix accumulation in the tubulointerstitial space ) , tubular atrophy , hyaline arteriosclerosis ( thickening/hardening of renal arteriole walls ) , and glomerulosclerosis ( scarring of glomerular capillaries and mesangial expansion ) [2 , 6] . Moreover , the cell senescence marker p16/INK4A increases with age in the human kidney , and correlates with interstitial fibrosis and glomerulosclerosis among older individuals [7] . These age-related changes in histopathology and cell senescence might underlie a decrease in kidney function with aging . However , the underlying molecular mechanisms that lead to age-associated changes in renal function and histopathology remain poorly characterized . The process of kidney aging has begun to be defined at a molecular level by identifying the global transcriptional changes that occur during aging [1 , 8] . In one study , DNA microarrays were used to profile gene expression in 74 kidneys aged 27 to 92 , and identified hundreds of genes that significantly change expression with age [1] . Notably , the gene expression signature for aging was correlated not only with the chronological age of the subjects , but also with the physiologic age of their kidneys , as determined by a histological score called the chronicity index . The chronicity index is an aggregate score of three histopathological measurements of renal aging in sections of renal tissue: renal interstitial fibrosis , tubular atrophy/hyaline arteriosclerosis and glomerulosclerosis . Importantly , the expression of kidney age-related genes correlated with the chronicity index adjusted for chronological age , such that an individual with a low chronicity index tended to have a renal gene expression profile characteristic of a younger individual , whereas an individual with a higher chronicity index tended to have a renal gene expression profiles characteristic of an older individual [1] . Thus , gene expression changes in the aging kidney not only reflect chronological age , but they correlate well with age-related changes in renal histopathology . These studies helped define how changes in gene expression may underlie deterioration of the structure and decline in function of the kidney in old age , but the upstream drivers of these changes in gene expression were previously unknown . With the availability of large genomic datasets from the ENCODE consortia , it is now possible to interrogate transcription factor ChIP-seq binding datasets to systematically identify candidate regulators of gene expression profiles [9–12] . In this study , we took an unbiased genomics approach to search for regulators of age-associated gene expression changes in the kidney , and identified the inflammation-related transcription factors STAT1 , STAT3 , and NFκB as top candidates . These three transcription factors become activated in old age and can account for a large fraction of the gene expression changes that occur during kidney aging . DNA polymorphisms in the two genes that encode subunits of the canonical NFκB transcription factor ( RELA and NFKB1 ) are associated with variation in kidney function and chronic kidney disease risk , providing genetic evidence that increasing NFκB activity in old age may play a causal role in age-related renal functional decline . This study highlights the dominant role played by STAT1 , STAT3 and NFκB in mediating transcriptional changes in the aging human kidney . Because these transcription factors mediate inflammatory responses , our results suggest that chronic inflammation underlies structural and functional decline of the kidney in old age .
To identify mechanisms underlying human kidney aging , we searched for transcription factors that bind nearby or within genes whose expression levels change with age in the kidney . In this study , we examined data for 961 ChIP-seq experiments in diverse human cell lines for 161 transcription factors to screen for datasets in which the set of genes bound by the transcription factor showed a significant overlap with the list of kidney age-related genes , an approach used previously for identifying regulators of aging in C . elegans ( see Methods ) ( Fig 1A ) [10] . Transcription factors in which the magnitude of overlap was >1 . 5-fold more than expected by chance and statistically significant after Bonferroni correction ( p < 5 x 10−5 ) were considered potential regulators of kidney aging , resulting in seven candidate regulators of the kidney aging transcriptome ( Fig 1B ) . The three transcription factors with the strongest enrichments for binding kidney age-related genes from ChIP-seq datasets were STAT1 ( 4 . 4-fold-enriched , p = 1 . 3 x 10−5 ) , STAT3 ( 2 . 1-fold enriched , p = 3 x 10−10 ) and NFκB ( 3 . 4-fold-enriched , p = 7 . 6 x 10−17 ) . STAT1 , STAT3 and NFκB were consistently enriched for binding kidney age-related genes in ChIP-seq experiments from different human cell lines ( S1 Table ) . To examine the specificity of STAT1 , STAT3 and NFκB for kidney aging , we asked whether these three transcription factors would show a strong enrichment for binding to diverse sets of genes from microarray expression data in the Gene Expression Omnibus ( GEO ) . We selected gene expression datasets from GEO to generate lists of significantly differentially-regulated genes from each expression dataset ( p < 0 . 001 ) . Two of the datasets were from microarray studies of human renal diseases ( diabetic kidney disease and nephrosclerosis ) and 18 datasets were selected at random from GEO microarray datasets for human Affymetrix U133 DNA chips . For each set of differentially-regulated genes , we determined whether the ChIP-seq binding peaks for STAT1 , STAT3 and NFκB showed a significant overlap ( defined as >1 . 5-fold enrichment , p < 5 x 10−5 and in the top 5% of most significantly-enriched ChIP-seq datasets ) . We found only two instances in which the ChIP-seq binding targets of STAT1 , STAT3 or NFκB showed a significant overlap; specifically , NFκB ChIP-seq targets from lymphoblastoid cell lines showed a strong overlap with genes that change expression in diabetic kidney disease and STAT3 ChIP-seq targets from MCF-10 breast cancer cell lines showed an overlap with genes that are differentially regulated upon knockdown of PGC-1α in melanoma cell lines ( S2 Table ) . The association between NFκB and diabetic kidney disease is consistent with previous work demonstrating that NFκB and its target genes are activated in human diabetic kidney disease [13] . These results indicate that STAT1 , STAT3 and NFκB do not generally show a significant enrichment for binding differentially-regulated genes from most microarray expression datasets . We compared the transcriptional profile of kidney aging from humans to a kidney aging transcriptional profile from rats [14] . We identified genes from the rat that were both orthologous to the 630 human kidney age-related genes and showed expression in the rat aging microarray experiment , resulting in a list of 427 genes . Next , we compared the age-related changes in expression of these 427 genes in human kidneys with the age-related changes in expression of the orthologous genes in rat kidneys , and found a significant correlation between their age-related changes in human and rat kidneys ( r = 0 . 46 , p < 10−5 ) ( S2 Fig ) . 114 of these 427 genes were significantly age-related in both human and rat kidneys ( 6 . 7-fold enrichment , p < 10−40 ) ( S3 Table ) . This result indicates that there is a common transcriptional signature of kidney aging in humans and rats . Since the human kidney samples were mainly obtained from diseased patients ( mostly individuals with renal cell carcinomas ) , whereas the rat samples were obtained from a normal aging cohort , the similarity between the human and rat kidney aging transcriptome indicates that the human gene expression pattern captures true age-related changes . Next , we asked whether STAT1 , STAT3 and NFκB bound to the common set of human and rat kidney aging genes; specifically , we calculated the overlap between the ChIP-seq binding peaks in the ENCODE datasets for these three transcription factors with the list of 114 kidney aging genes shared between human and rat . The ChIP-seq binding targets for STAT1 , STAT3 and NFκB were significantly enriched for binding these 114 kidney age-related genes , suggesting that these inflammatory transcription factors may contribute to age-related gene expression changes in both humans and rats ( S4 Table ) . STAT1 and STAT3 belong to the highly conserved JAK/STAT family of transcription factors . The activation and nuclear localization of STAT transcription factors is primarily regulated by tyrosine phosphorylation , dimerization and translocation to the nucleus in response to specific cytokines or growth factors [15 , 16] . Inducers of the STAT transcription factors include inflammatory cytokines , such as interleukin-6 ( IL-6 ) ( primarily an activator of STAT3 ) or interferon gamma ( IFNγ ) ( primarily an activator of STAT1 ) [16 , 17] . Similarly , the canonical NFκB transcription factor , a heterodimer of p50 ( encoded by the NFKB1 gene ) and p65/RelA ( encoded by the RELA gene ) can be activated by inflammatory cytokines , such as tumor necrosis factor alpha ( TNFα ) and interleukin-1 beta ( IL-1ß ) , as well as DNA damage , lipopolysaccharide , or reactive oxygen species [18 , 19] . To test if the three candidate regulators ( STAT1 , STAT3 and NFκB ) are responsible for driving gene expression changes during kidney aging , we assessed the expression levels and activity of these transcription factors in young and old human kidney tissues . We first examined the age-associated changes in the mRNA expression levels of the genes encoding these three transcription factors , using previously published microarray gene expression data [1] . We found a significant age-related increase in the mRNA expression levels of STAT1 ( 1 . 23-fold increase , p = 0 . 006 , Student’s t-test ) and STAT3 ( 1 . 2-fold increase , p = 0 . 008 , Student’s t-test ) in old versus young kidneys , but no significant change in RELA mRNA expression levels with age ( Fig 2A ) . We next performed immunohistochemistry ( IHC ) experiments to assess the localization patterns , expression levels and activities of STAT1 , STAT3 and NFκB ( RelA ) protein in paraffin sections of renal cortex tissues from young ( 25–44 years of age ) and old ( 66–85 years of age ) individuals . STAT1 , STAT3 and RelA were predominantly localized to the cytoplasm with sporadic nuclear staining in a small fraction of tubular epithelial and glomerular cells . Expression of these three transcription factors was observed in multiple renal cell types , including tubular epithelial , glomerular and interstitial cells ( Fig 2B ) . We found a significant increase in overall STAT1 immunoreactivity with age ( p = 0 . 02 , Mann-Whitney U test ) , but no significant change in STAT3 or RelA immunoreactivity with age . To assess differences in transcription factor activity between old and young kidneys , we used antibodies that recognize activated ( phosphorylated ) forms of these transcription factors ( herein abbreviated as pSTAT1 , pSTAT3 and pRelA ) . We observed significantly higher nuclear levels of pSTAT1 ( p = 0 . 002 , Mann-Whitney U test ) , pSTAT3 ( p = 0 . 003 , Mann-Whitney U test ) and pRelA ( p = 0 . 006 , Mann-Whitney U test ) in old versus young renal cortex ( Fig 2B ) . Activation of these three transcription factors was observed in multiple epithelial cell types in aged kidneys . Immunoreactivity for pSTAT1 was restricted to subsets of tubules , a few cells within glomeruli , and Bowman’s capsule epithelial cells . The tissue expression pattern for pSTAT3 in older renal cortex tissues was somewhat broader than that of pSTAT1 , and included a larger subset of tubular epithelial cells , glomerular cells and interstitial cells . Expression of pRelA in older kidneys spanned most regions of the renal cortex , including the majority of renal tubular epithelial cells , subsets of glomerular cells and interstitial cells . To test if STAT1 , STAT3 and NFκB activation drive transcriptional changes of their target genes during kidney aging , we first sought to identify the set of target genes directly regulated by these transcription factors in human renal tubular epithelial cells ( HK-2 cells ) . Genes were considered to be direct targets of each transcription factor if they were directly bound to the transcription factor in ChIP-seq datasets [11] and if their expression was responsive to transcription factor activation by inflammatory cytokine stimulation . We used IFNγ to activate STAT1 signaling , IL-6 to activate STAT3 signaling and TNFα to activate NFκB signaling in HK-2 cells . IFNγ predominantly regulates gene expression via STAT1 activation , as knockdown of STAT1 in IFNγ-stimulated cells represses the vast majority of IFNγ-induced genes [15] . Likewise , IL-6 regulates gene expression predominantly through STAT3 , as 86% of IL-6-induced transcriptional changes were suppressed by STAT3 inhibition of IL-6 treated HK-2 cells using the STAT3-specific inhibitor drug S3I-201 ( p < 10−5 , binomial test ) ( S3 Fig ) . Finally , about 90% of the transcripts that increase expression in response to TNFα in human cells are NFκB-dependent [17] We used DNA microarrays to profile changes in gene expression following stimulation of HK-2 cells with the inflammatory cytokines IFNγ , IL-6 or TNFα . To define a set of STAT1 , STAT3 or NFκB direct target genes we combined transcription factor binding data from ChIP-seq experiments with microarray expression data to identify genes that are bound by the transcription factor and differentially regulated by cytokine stimulation of HK-2 cells ( see Methods ) . We thereby identified 40 STAT1 direct target genes , 43 STAT3 direct targets , and 43 NFκB direct targets ( Fig 3; S5 Table ) . Next , we compared the expression changes of STAT1 , STAT3 and NFκB regulated-targets in HK-2 cells with their expression changes during kidney aging . Expression changes for the direct targets of STAT1 following IFNγ stimulation largely recapitulated their expression changes during aging ( Fig 3A ) . Notably , the majority of STAT1 direct targets that increased expression upon IFNγ stimulation also increased expression during kidney aging ( 84% concordance , p < 10−4 , Binomial test ) . These STAT1-induced direct targets include genes involved in the regulation of apoptosis ( e . g . PML , BAK1 , MX1 ) , antigen presentation ( e . g . HLA-E , TAP1 ) and members of the JAK-STAT signaling pathway ( e . g . SOCS3 , STAT1 , STAT2 , STAT3 ) . For STAT3 , we profiled changes in gene expression upon stimulation of HK-2 cells with IL-6 ( for STAT3 activation ) or STAT3-specific inhibition by the drug S3I-201 following IL-6 treatment . The expression profile of the 43 STAT3 direct targets following IL-6 stimulation was correlated with the changes in their expression during aging ( r = 0 . 40 , p = 0 . 01 ) ( Fig 3B ) . The majority of STAT3 targets that were induced by IL-6 were also induced during kidney aging ( 84% concordance , p < 0 . 001 , Binomial test ) . To inhibit STAT3 activation , we stimulated HK-2 cells with IL-6 followed by addition of the STAT3 inhibitor S3I-201 . Gene expression changes for the STAT3 direct targets following STAT3 inhibition were strongly anti-correlated with their changes in gene expression during kidney aging ( r = -0 . 51 , p = 0 . 003 ) ( Fig 3B ) . Thus , activation of STAT3 led to a transcriptional response characteristic of kidney aging , while inhibition of STAT3 elicited a transcriptional program characteristic of a younger kidney . The set of 43 STAT3 direct targets include genes known to be involved in the regulation of cell proliferation ( e . g . MYC ) , metabolic functions ( e . g . NNMT , NAMPT ) and regulation of apoptosis ( e . g . BIRC3 , BCL6 , TNFRSF1A ) . Expression changes of the direct target genes of NFκB following TNFα treatment were strongly correlated with their gene expression changes in the aging kidney ( r = 0 . 54 , p < 10−4 ) ( Fig 3C ) . 35 out of the 40 NFκB target genes that were induced by TNFα were induced during kidney aging ( 88% concordance , p < 10−5 , binomial test ) . The set of NFκB-direct targets include mediators of innate immune responses ( e . g . LTB , ICAM1 ) , apoptosis ( e . g . BIRC3 , TRAF3 ) and feedback regulators of NFκB signaling ( e . g . NFKBIA , NFKBIE , NFKBID ) . The direct targets of STAT1 , STAT3 and NFκB were largely non-overlapping with each other , suggesting that combinatorial activation of these transcription factors might have additive effects on the aging gene expression profile ( S4 Fig ) . To investigate whether simultaneous activation of STAT1 , STAT3 and NFκB might have additive effects in promoting the aging transcriptional phenotype , we used DNA microarrays to profile changes in gene expression following simultaneous stimulation of HK-2 cells with each possible combination of IL-6 , IFNγ and TNFα . We then compared the resulting changes in gene expression for the direct targets of STAT1 , STAT3 or NFκB in HK-2 cells with the changes in their expression during kidney aging . Generally , we found that stimulation of HK-2 cells with two or more of these inflammatory cytokines recapitulated more of the kidney age-induced transcriptional profile than IL-6 , IFNγ , or TNFα alone ( S5 Fig ) . There was a particularly strong correlation in the gene expression profile of these genes following simultaneous treatment with the three inflammatory cytokines and their expression behavior during kidney aging ( r = 0 . 51 , p < 10−5 ) . Notably , the vast majority of direct targets of STAT1 , STAT3 and NFκB that were induced by combined treatment with IFNγ , IL-6 , and TNFα were also induced during kidney aging ( 90% concordance , p < 10−9 , binomial test ) . To go beyond the direct targets of STAT1 , STAT3 and NFκB , we wanted to ask whether the general transcriptional responses to the IFNγ , IL-6 , or TNFα resembled age-associated gene expression changes in the kidney . These target genes may include both direct and indirect targets of the three transcription factors . For each cytokine , we identified genes that were significantly differentially regulated ( p < 0 . 001 ) in our DNA microarray experiments upon cytokine stimulation of HK-2 cells ( see Methods ) . This analysis identified 48 IFNγ-regulated genes , 32 IL-6-regulated genes and 230 TNFα-regulated genes ( S5 Table ) . For these cytokine-regulated genes , the kidney age-related gene expression changes were significantly correlated with their transcriptional changes in response to IFNγ ( r = 0 . 64 , p < 10−5 ) , IL-6 ( r = 0 . 36 , p = 0 . 02 ) and TNFα ( r = 0 . 46 , p < 10−10 ) . IFNγ induced a number of canonical interferon responsive genes ( IRF1 , GBP1 , GBP2 ) , and genes involved in innate immune responses ( IL18BP , TNFRSF1A , ICAM1 ) . IL-6 induced genes involved in feedback regulation of STAT3 signaling ( e . g . SOCS3 , STAT3 ) , as well as inflammation-related transcription factor genes ( e . g . CEBPD , BCL6 ) . TNFα induced several inflammatory cytokine genes ( e . g . IL6 , IL32 , IL23A , IL8 ) , chemokine genes ( e . g . CCL2 , CXCL2 , CCL20 ) , complement factors ( e . g . C3 , CFI ) , and regulators of apoptosis ( e . g . FAS , TNFAIP3 , BIRC3 ) . These results suggest that IFNγ , IL6 and TNFα may contribute to activation of the inflammatory response in the aging kidney . As a control , we examined whether the overlap between the kidney aging transcriptome and transcriptional responses to IL-6 , IFNγ or TNFα are specific to these cytokines or whether other cytokines tend to show a similar effect . We obtained published gene expression datasets from GEO for the transcriptional response of human cells to ten different cytokines and growth factors [20–30] . For each gene expression dataset , we then identified genes that showed a significant change in expression ( p < 0 . 001 ) following cytokine or growth factor stimulation . Finally , we calculated the correlations between the gene expression changes following cytokine/growth factor treatment and their expression changes during kidney aging . The IL-6 , IFNγ and TNFα transcriptional responses showed the most significant correlations with kidney aging from this list ( S6 Fig ) . Of the ten other cytokines , the transcriptional responses to eight did not significantly correlate with the kidney aging gene expression profile and the transcriptional responses to IL-1ß and VEGFA showed a moderate correlation ( S6 Fig ) . The association between kidney aging and the IL-1ß transcriptional response might be expected given that IL-1ß activates NFκB , similar to TNFα [18] . This analysis indicates that the kidney aging transcriptome shows a strong overlap with the transcriptional responses evoked by IFNγ , IL-6 , and TNFα , but not with the transcriptional responses evoked by various other cytokines . A common histopathologic feature of kidney aging is tubulointerstitial fibrosis , a progressive scarring of the kidney , characterized by accumulation of extracellular matrix in the renal interstitial space , and an increased abundance of myofibroblasts , the primary mesenchymal cell type responsible for increased collagen deposition in fibrotic kidneys [31] . One cellular mechanism of interstitial fibrosis is epithelial to mesenchymal transition ( EMT ) , whereby tubular epithelial cells acquire characteristics of fibroblasts and myofibroblasts in response to injury or inflammatory stimuli [31] . Microarray analysis of differentially-regulated genes following TNFα stimulation of HK-2 cells showed increased expression of several genes characteristically expressed in fibroblasts and myofibroblasts , including the EMT transcription factor TWIST1 and the mesenchymal marker vimentin ( S6 Table ) . These microarray results suggest that TNFα might initiate a mesenchymal differentiation program in HK-2 cells . Previous work has shown that NFκB can drive EMT in the context of cancer metastasis . Specifically , TNFα stimulation or overexpression of RelA is sufficient to promote EMT of carcinoma cells [32–34] . We extended these observations by asking whether TNFα could promote a mesenchymal phenotype in HK-2 cells . As expected , TNFα treatment resulted in increased levels of nuclear pRelA , indicating NFκB activation ( Fig 4A ) . We found that prolonged TNFα treatment caused HK-2 cells to acquire an elongated morphology characteristic of mesenchymal cells ( Fig 4B ) , as well as increased protein expression of the mesenchymal marker vimentin ( Fig 4A ) , the expression of which also increases significantly with age ( 1 . 43-fold increase , p < 10−4 ) ( Fig 4C ) . These findings indicate that TNFα is sufficient to induce a mesenchymal transition in HK-2 cells , consistent with the results of a previous study [35] . Since renal fibrosis is associated with an increase in extracellular collagen deposition , we asked whether collagen genes tended to be induced by TNFα , and found that the majority of detectable collagen transcripts increased expression following TNFα stimulation and similarly increased during kidney aging ( Fig 4D and S6 Table ) . Taken together , our results suggest that NFκB activation may contribute to the development of interstitial fibrosis in the aging kidney by promoting mesenchymal differentiation and fibrotic gene expression . One possibility is that age-associated activation of STAT1 , STAT3 and NFκB is caused by a common mechanism ( e . g . chronic inflammation ) , in which case their activities should co-vary in individuals . That is , in individuals of the same age , the levels of STAT1 , STAT3 and NFκB transcription factor activity might fluctuate in a coordinated fashion , suggesting that they respond to individual differences in a common upstream mechanism . In order to ask whether STAT1 , STAT3 and NFκB target genes are co-regulated in individual kidney samples , we first devised a measurement to reflect the activity of these transcription factors in different individuals using DNA microarray expression data from 73 kidneys . We used the expression levels of the direct target genes for STAT1 , STAT3 and NFκB , respectively , as a proxy for their activity in individual kidney samples . For each renal cortex sample , we calculated a normalized z-score indicating whether expression of the target genes in that individual kidney was higher or lower than the average score for that age . We then compared the activity scores for the three transcription factors to each other , and found that there was a strong correlation in the age-adjusted activity levels of STAT1 , STAT3 and NFκB in different individuals ( Fig 5A ) . This result shows that the activities of these three transcription factors co-vary in individuals , such that those with relatively high expression of NFκB target genes also tend to have relatively high expression of STAT1 and STAT3 target genes for their age . These results suggest that there might be a common mechanism underlying the coordinate regulation of these three transcription factors . We hypothesized that macrophages might contribute to coordinated activation of STAT1 , STAT3 and NFκB , since they secrete high levels of many inflammatory cytokines [36] . As preliminary evidence that macrophage abundance increases with age , we found that many transcripts specifically expressed in monocytes and macrophages ( e . g . CD14 , CD163 ) increase expression during kidney aging by analyzing microarray expression data from [1] . To examine how macrophage abundance changes with age in vivo , we performed immunohistochemistry using antibodies to CD163 , a specific marker for macrophages [37] . We found a 2 . 1-fold higher abundance of interstitial macrophages in renal cortex sections from old versus young individuals ( p = 0 . 006; Fig 5B ) . To test whether individual variation in macrophage abundance correlates with variation in the activity of STAT1 , STAT3 and NFκB , we first used the expression of three macrophage-specific transcripts ( CD163 , CD14 , TYROBP ) in the kidney DNA microarray data as an estimate of macrophage abundance in each renal cortex sample ( see Methods ) . If macrophage infiltration contributes to the age-related increase in the activities of STAT1 , STAT3 and NFκB , then one would expect that there would be a correlation between macrophage abundance and transcription factor activity in individual kidney samples , normalized for their ages . That is , among kidneys of the same age , those with higher relative macrophage abundance should also show higher levels of STAT1 , STAT3 and NFκB activity , and vice versa . For each individual , we used linear regression to calculate whether that individual kidney sample showed high or low levels of macrophage-specific gene expression compared to the expected levels for their age . We then compared the estimated abundance of macrophages to the levels of activation of STAT1 , STAT3 and NFκB in 73 individuals , adjusting for chronological age . We generated a heat map showing the estimated abundance of macrophages and transcription factor activity in the renal cortex , normalized for age . The heat map displays the correlation between estimated macrophage abundance with transcription factor activation as a dendrogram . We found that macrophage abundance was highly correlated with the activity of these three inflammatory transcription factors ( r = 0 . 66–0 . 81 , p < 10−6 ) ( Fig 5A ) . The magnitude of this correlation suggests that macrophage abundance explains 40–65% of the variance in the activity of the three inflammatory transcription factors . In contrast , the expression of genes characteristically expressed in other immune cell types ( i . e . T and B cells , natural killer cells , plasma cells , neutrophils ) were not strongly correlated with the activity of STAT1 , STAT3 , NFκB , or with estimated macrophage abundance ( S7 Fig ) . To further examine the association between renal macrophage infiltration and activation of these transcription factors in vivo , we performed double-label immunofluorescence experiments to examine the correlation between interstitial macrophage abundance ( CD163 ) and levels of nuclear pSTAT3 and pRelA in human kidney sections . We found that interstitial CD163+ macrophage infiltration was significantly correlated with pSTAT3 and pRelA nuclear immunoreactivity in epithelial compartments of the renal cortex ( Fig 5C ) . Among the young kidney sections , the individual with the highest level of macrophage infiltration also had the highest level of STAT3 activation . Conversely , from the elderly , the individual kidney with the lowest burden of interstitial macrophages also had the lowest level of STAT3 activation ( Fig 4C , r = 0 . 77 , p < 0 . 001 ) . We obtained similar associations between CD163+ macrophage abundance and NFκB activation ( r = 0 . 83 , p < 0 . 001 ) ( Fig 5C ) . These results indicate that macrophage abundance is correlated with the activity of these inflammatory transcription factors in renal epithelial compartments , independent of chronological age . Since activated macrophages are a source of inflammatory cytokines , we hypothesized that the transcriptional response evoked by macrophage-secreted factors might recapitulate gene expression changes associated with kidney aging [36] . To explore this , we analyzed gene expression changes in response to conditioned media from macrophages using data from two DNA microarray studies that characterized the transcriptional responses of human cells ( endometrial stromal cells and adipocytes ) to macrophage-conditioned media [38 , 39] . We identified a set of 77 genes that showed a concordant pattern of differential expression ( p < 0 . 001 in both studies ) ( S7 Table ) in response to macrophage-conditioned media , thereby defining a core gene expression signature for the response of human cells to macrophage-conditioned media . The transcriptional response to macrophage-conditioned media was strongly correlated with the gene expression changes that occur during kidney aging ( r = 0 . 42 , p < 10−4 ) ( Fig 4D ) . This analysis indicates that macrophage-secreted signals are sufficient to evoke gene expression changes similar to those that occur during kidney aging . Given our observation that STAT1 , STAT3 and NFκB contribute to the aging transcriptional program , we used the genes for these transcription factors as candidates to find out if they are associated with either kidney function or renal disease susceptibility . Specifically , we asked whether common single nucleotide polymorphisms ( SNPs ) in these transcription factor genes are associated with individual differences in estimated glomerular filtration rate ( eGFR ) or chronic kidney disease risk in the human population . We used publicly-available data from a large genome-wide association study of kidney function and chronic kidney disease that included 67 , 093 Caucasian individuals to test whether SNPs in STAT1 , STAT3 , or genes encoding the canonical NFκB transcription factor ( RELA or NFKB1 ) are associated with either eGFR or chronic kidney disease [40] . We selected 217 SNPs within these four genes and queried the p-values for their associations with eGFR [40] . We identified lead SNPs in NFKB1 ( rs12509403 ) and RELA ( rs11820062 ) that showed significant associations with eGFR after Bonferroni correction for multiple testing ( Table 1 and Fig 6 ) . The SNP rs11820062 in RELA also showed a significant association with chronic kidney disease susceptibility ( OR = 1 . 088 , p = 8 . 0 x 10−5 ) ( Table 1 ) . rs11820062 in RELA is moderately linked to rs4014195 ( R2 = 0 . 44 ) , an intergenic SNP that is located approximately 80 kilobases upstream of the RELA transcription start site ( Fig 6A ) . rs4014195 has been previously associated with eGFR ( p = 3 x 10−8 ) at genome-wide level statistical significance [40] . One possibility is that the phenotypic effect of rs4014195 on kidney function is mediated by changes in RelA expression . Alternatively , the phenotypic effect of rs4014195 might be due to changes in the activity of another nearby gene or several genes in its LD region . Neither rs12509403 in NFKB1 nor rs11820062 in RELA are linked to common SNPs that cause protein-coding changes . To ask whether these genetic variants may result in gene expression changes , we queried a large public dataset of expression quantitative trait loci ( eQTLs ) in peripheral blood [41] . We found that rs12509403 in NFKB1 is associated with significant differences in mRNA expression of NFKB1 ( p = 1 . 03 x 10−7 ) . Specifically , the C allele is associated with lower NFKB1 expression , while the T allele is associated with higher NFKB1 expression . The lower expression rs11820062 ( C ) allele is associated with a lower eGFR . This effect of the risk allele in NFKB1 ( encoding p50 ) is consistent with its known role in antagonizing the activity of the NFκB complex . The p50 subunit is found either as part of the RelA-p50 NFκB heterodimer or in p50-p50 homodimers . The p50-p50 homodimer binds and represses NFκB target genes by preventing their activation by RelA-containing complexes [18 , 19] . The genetic association result for NFKB1 in humans shows a weak effect in the same direction as the null phenotype in mice; specifically , the human low-expressing NFKB1 allele is associated with decreased kidney function and the NFKB1 knockout phenotype in mice is characterized by increased RelA target gene expression and higher levels of chronic inflammation [42] . We examined SNPs linked to rs12509403 in order to identify candidates that might account for the variation in gene expression in NFKB1 . For every SNP that is strongly linked to the lead SNP rs12509403 ( R2 > 0 . 8 ) , we examined data from ENCODE and asked whether the SNP resides within an H3K4me1 peak ( often marks promoters ) , H3K27 acetylation peak ( often marks active enhancers ) , a DNase I hypersensitivity cluster ( marks accessible chromatin ) or affects binding to a specific transcription factor . rs4640855 is linked to rs12509403 and falls within a ChIP-seq binding site for the Fos transcription factor , a DNase I hypersensitivity peak , and a H3K4me1 peak ( S8 Fig and S8 Table ) . rs1598859 and rs3774963 are also linked to rs12509403 and fall within a DNase I hypersensitivity peak , an H3K4me1 peak and an H3K27 acetylation peak ( S8 Fig and S8 Table ) . These three SNPs are candidates for being responsible for changes in the expression level of NFKB1 .
We performed a high-throughput search for transcription factors that control gene expression changes during kidney aging . Out of 161 transcription factors screened , the top three hits from our analysis were the inflammation-associated transcription factors STAT1 , STAT3 and NFκB , suggesting that they play a relatively important role in regulating age-associated gene expression changes in the kidney . We showed that STAT1 , STAT3 and NFκB increase with age in the kidney , and that these increases can partially recapitulate the kidney aging transcriptome . A common feature of these transcription factors is that they mediate inflammatory responses . Thus , our results suggest a model in which the activity of STAT1 , STAT3 and NFκB increase with age in the kidney , leading to transcriptional cascades and chronic inflammation in old age . The regions of the kidney where STAT1 , STAT3 and NFκB become activated in old age provide clues as to how these transcription factors are related to specific renal aging phenotypes . STAT1 was most strongly activated along Bowman’s capsule epithelial cells and in some cells along the periphery of glomeruli , a region where small capillaries become scarred in older kidneys ( glomerulosclerosis ) . STAT3 activity was increased in glomerular cells , subsets of tubules and interstitial stromal cells in old age . Finally , NFκB was activated in most cells of glomerular and tubulointerstitial compartments in older renal cortex tissues . In summary , the regions showing activation of STAT1 , STAT3 and NFκB during aging suggests a possible role for these transcription factors in the progression of glomerulosclerosis and renal interstitial fibrosis during aging . We showed that activation of STAT1 , STAT3 and NFκB by inflammatory cytokines in HK-2 cells caused transcriptional changes in their direct targets that are similar to their expression changes during kidney aging . We restricted our analysis to direct transcription factor targets in order to capture the primary effects of these transcription factors . It is important to note that STAT1 and STAT3 share several of the same direct targets ( e . g . STAT3 , BCL6 , NNMT , TNFRSF1A ) . Moreover , STAT1 and STAT3 can compete for binding at the same DNA sequence motifs , or they can regulate target gene expression as STAT1-STAT3 heterodimers [43] . Hence , there may be redundancy or cooperativity in the roles of STAT1 and STAT3 in regulating transcriptional changes during kidney aging . By activating STAT1 , STAT3 and NFκB simultaneously , we were able to cause transcriptional changes that recapitulated a significant portion of the kidney aging expression profile . Specifically , the correlation between the transcriptional changes caused by activation of all three transcription factors together and aging was r = 0 . 51 ( p < 10−5 ) . The magnitude of this correlation , as well as the strong ChIP-seq binding enrichments for these transcription factors at kidney age-related genes indicates that the combined effects of STAT1 , STAT3 and NFκB target gene activation play an important role in regulating gene expression changes in the aging kidney . Is increased activity of STAT1 , STAT3 and NFκB detrimental to the pathophysiology of older kidneys ? Several studies suggest that increased activity of JAK/STAT and NFκB transcription factors can have deleterious effects on the kidney in response to renal injury or inflammatory renal disease . In mice with lupus nephritis ( an inflammatory kidney disease ) , STAT1 becomes activated and blockade of STAT1 activity reduces macrophage infiltration and improves renal function [44 , 45] . Similarly , in a mouse model of obstructive kidney injury , STAT3 is activated and inhibition of STAT3 with the drug S3I-201 attenuates interstitial fibrosis and immune cell infiltration in the injured kidney [46] . In a mouse model of diabetic kidney disease ( the most common form of human chronic kidney disease ) , mice that are heterozygous for a STAT3 loss of function mutation have reduced interstitial fibrosis and inflammation , and improved renal function compared to diabetic mice with wild-type STAT3 [47] . Other studies have reported that inhibition of NFκB activity has a protective effect in rodent models of renal injury and renal aging [48 , 49] . While most of these studies were not performed in the context of aging , these rodent models of renal injury and disease recapitulate many of the histopathologic features of human kidney aging , such as interstitial fibrosis and inflammation . Therefore , the results from the renal disease models suggest that elevated activity of STAT1 , STAT3 and NFκB may have a deleterious effect on kidney physiology and function in old age . Moreover , previous work has shown that NFκB plays a multifaceted role in promoting aging-related changes in several tissues [6 , 42 , 50–52] . The inflammatory cytokines IFNγ , IL-6 and TNFα are upstream activators of STAT1 , STAT3 and NFκB , respectively . The systemic circulating levels of IL-6 and TNFα increase with age and the production of IFNγ by subsets of lymphocytes is increased in elderly individuals [53–55] . These observations suggest that these three cytokines may be partially responsible for increased activity of STAT1 , STAT3 and NFκB in the kidney in old age . However , whether IFNγ , IL-6 or TNFα are the main factors responsible for activating STAT1 , STAT3 and NFκB in vivo during kidney aging remains unclear , since other signals can also activate these three transcription factors . For example STAT1 can be activated by type I interferon , STAT3 can be activated by growth factors ( e . g . EGF , LIF ) and NFκB can be activated by reactive oxygen species , infection and other inflammatory cytokines , including IL-1β [18 , 19] . Further studies in vivo should illuminate which of these upstream signals are responsible for activation of STAT1 , STAT3 and NFκB in the aging kidney . Macrophages are known to secrete a variety of inflammatory cytokines , including IL-6 , IFNγ and TNFα [36] . Three observations suggest that renal macrophage infiltration might contribute to the activation of STAT1 , STAT3 and NFκB in the kidney . First , macrophage infiltration increases with age in the human kidney . Second , the general transcriptional response of human cell lines to macrophage-conditioned media recapitulates gene expression changes that occur during kidney aging . Third , variation in the activity of STAT1 , STAT3 and NFκB are correlated with variation in macrophage abundance between individuals , independent of chronological age . This correlation suggests that there is a common underlying mechanism linking increased macrophage infiltration with the activity of these three transcription factors in the kidney epithelium , or that macrophages themselves may potentiate the activation of STAT1 , STAT3 and NFκB in the renal parenchyma . Nevertheless , it remains unclear whether macrophage infiltration is the primary contributor to inflammatory gene expression changes in the epithelial compartments during kidney aging , and whether macrophage infiltration contributes to functional decline of the kidney . In mice , blocking macrophage infiltration via inhibition of CCR2 ( the receptor for CCL2/monocyte chemotactic protein-1 ) reduces renal fibrosis and inflammation and improves renal function in diabetic kidney disease , suggesting that infiltrating macrophages might play a causative role in the progression of renal fibrosis [56] . In addition to macrophages , other tissues and cell types have been suggested to promote chronic inflammation in the aging kidney , including senescent cells and peripheral adipose tissue [57] . Since STAT1 , STAT3 and NFκB regulate gene expression changes during aging , we wanted to examine the functional role of these transcription factors in the human kidney . That is , does genetic variation in the genes encoding these transcription factors influence kidney function or chronic kidney disease susceptibility ? Glomerular filtration rate varies considerably in individuals of the same age . In old age , some individuals maintain relatively high levels of kidney function , while others suffer from chronic kidney disease [3] . We used data from a large genome-wide association study [40] to examine whether common SNPs in any of the genes encoding the kidney aging transcription factors ( STAT1 , STAT3 and the canonical NFκB complex genes RELA and NFKB1 ) show an association with kidney function or chronic kidney disease . We found two independent DNA variants in the NFκB transcription factor genes RELA and NFKB1 ( rs11820062 and rs12509403 , respectively ) that associate with kidney function and chronic kidney disease susceptibility in the human population . rs12509403 in NFKB1 is also associated with variation in NFKB1 gene expression [58] , suggesting that variation in expression of NFKB1 may explain the phenotypic effect of this SNP on kidney function . Our findings provide genetic evidence that variation in the genes encoding the canonical NFκB transcription factor are associated with differences in kidney function and chronic kidney disease susceptibility in the human population . Furthermore , the genetic association between NFκB and age-related renal phenotypes suggests that the observed increase in NFκB activity during kidney aging may contribute to the age-related decline in renal function . In this study , we identified three inflammation-induced transcription factors that drive transcriptional cascades in the aging kidney . Our work adds to a growing body of evidence that chronic inflammation is a contributing factor to aging phenotypes in human tissues . Future studies should provide additional insights about how STAT1 , STAT3 and NFκB affect renal function and age-related phenotypes in vivo , as well as the primary upstream signals responsible for their activation in the aging kidney .
961 ChIP-seq data sets for 161 transcription factors ( on human genome version hg19 ) were obtained from the ENCODE Consortium as of 07/13/12 ( http://genome . ucsc . edu/ENCODE/downloads . html ) . All binding sites with q-value or p-value < 10−5 were considered , and ChIP-seq peaks less than 20 base pairs long were discarded . Binding sites were then associated with ENSEMBL-annotated hg19 transcripts ( ensGene , downloaded 12/9/13 ) if the position of maximum read density within the binding site was located within 5 kb upstream or 2 kb downstream of the annotated transcription start site . Prior work has shown that a parameter termed complexity affects whether or not a ChIP-seq binding peak for a specific transcription factor is likely to identify a gene whose expression is responsive to knockdown of that transcription factor [10] . For most transcription factors , binding sites vary from low complexity sites ( those bound by a small number of transcription factors ) to high complexity sites ( those bound by many transcription factors ) . Sites bound by a large number of transcription factors are not generally responsive to changes in just one transcription factor . We calculated the complexity of every ChIP-seq binding site , defined as the number of transcription factors that were found to have a significant binding site that overlaps that region . Overlapping binding sites for the same transcription factor observed in multiple cell lines or replicate experiments were only counted once . We performed a pilot analysis to investigate how binding complexity might affect the selection of transcription factors that bind to the age-related kidney genes . The list of kidney age-related genes was obtained from supplemental data published in [1] , which originally identified age-regulated probesets corresponding to 630 genes used for the analysis in this paper . We compared the overlap between ChIP-seq binding peaks and age-related kidney genes using all of the binding peaks , or only the binding peaks with <50% complexity ( i . e . bound by 81 or fewer transcription factors ) . To focus our analysis on transcription factors that act within the renal parenchyma , we selected transcription factors that showed detectable expression in epithelial compartments using IHC data from the human protein atlas . We compared the degree of overlap between the list of ChIP-seq targets to the list of age-regulated genes . The results for the screen using the binding sites with <50% complexity is shown in Fig 1B , and the results using all of the transcription factor binding sites are shown in S9 Fig . The two screens returned similar lists of transcription factors enriched for binding to age-related kidney genes; specifically , STAT1 , NFκB and STAT3 were identified consistently . For this paper , we selected transcription factor binding targets with a binding complexity <50% . For enrichment calculations , the set of all genes present on the U133A and B microarrays were used as a background set . Significance of overlap was determined by Fisher’s Exact test , with Chi-Square approximation where appropriate ( all values greater than 5 ) . To restrict our analysis to bona fide transcription factors , we excluded non-specific DNA-binding machinery , such as factors associated with RNA polymerase II and components of basal transcriptional machinery ( e . g . p300 , CTCF , RAD21 , TAF1 ) . Finally , we selected only transcription factors that showed detectable staining in epithelial compartments of the kidney ( tubules or glomeruli ) in IHC experiments generated by the human protein atlas ( http://www . proteinatlas . org/ ) . We thereby removed transcription factors that are specifically expressed in immune cells ( i . e . PU1 , EBF1 , IKZF1 , BCL11A , BCLAF1 ) , which would otherwise appear enriched in our screen . Transcription factor ChIP-seq experiments with a Chi-square or Fisher’s exact p-value of less than 5 x 10−5 ( significant after Bonferroni correction for all ChIP-seq experiments tested ) and a >1 . 5 fold enrichment were considered statistically significant . We obtained paraffin-embedded normal human kidney samples from the Stanford Department of Surgical Pathology . Samples were extracted from tumor-free normal tissue in patients who underwent nephrectomy for renal tumors . Tissues were sectioned onto slides for hematoxylin and eosin staining and immunostaining experiments . Slides containing sections of paraffin-embedded kidney tissues were deparaffinized in three changes of xylene . Antigen retrieval was performed in a pressure cooker using either tris-EDTA ( pH 9 . 0 ) for pSTAT3 , pRelA and CD163 antibodies or in citrate buffer ( pH 6 . 0 ) for STAT3 and NFκB p65/RelA antibodies . Blocking and washing steps were performed according to the manufacturer’s protocol using the Expose detection IHC kit ( Abcam , Cambridge , MA ) . The primary antibodies used were rabbit polyclonal STAT3 ( 1:400 , Cell Signaling , Danvers , MA ) , rabbit monoclonal to phosphoTyr705 STAT3 ( 1:20 , Cell Signaling ) , rabbit polyclonal NFκB p65/RelA ( 1:80 , Cell Signaling ) , rabbit polyclonal phosphoSer536 NFκB p65/RelA ( 1:40 , Novus Biologicals , Littleton , CO ) , rabbit polyclonal to STAT1 ( 1:80 , Abcam ) , rabbit polyclonal pSTATTyr701 ( 1:60 , Sigma-Aldrich , St . Louis , MO ) . The appropriate horseradish peroxidase-conjugated secondary antibodies were used for chromogenic signal detection with 1 , 2-diaminobenzidene in buffered substrate ( Expose detection IHC kit , Abcam ) . Slides were counterstained with hematoxylin . Semi-quantitative scoring of IHC stains was performed blinded to chronological age , using a 0–5 point relative scoring scheme . Scores reflect both the signal intensity and the percentage of positively-stained cells from three renal cortical fields at 20X magnification , similar to previously described composite scoring schemes [59 , 60] . For pSTAT1 , pSTAT3 and pRelA , only nuclear staining was scored as a marker for transcription factor activation . Statistical differences in IHC staining scores between the young and old renal cortex tissues were calculated using the Mann-Whitney U test ( one-tailed ) , testing the hypothesis that transcription factor activity increases with age . For double-label immunofluorescence staining of paraffin-embedded tissue sections , slides were blocked for 1 hour in blocking buffer ( 5% bovine serum albumin/0 . 5% Triton in PBS ) , rinsed in PBS and incubated overnight at 4°C in primary antibodies . For co-immunofluorescence staining of HK-2 cells , cells were fixed in 4% paraformaldehyde and blocked for 1 hour in blocking buffer ( 5% bovine serum albumin/0 . 5% Triton in PBS ) , rinsed in PBS , and incubated for 1 hour in primary antibodies at room temperature . Primary antibodies used were mouse monoclonal anti-vimentin ( Abcam ) and rabbit polyclonal phosphoSer536 NFκB p65/RelA ( 1:40 , Novus Biologicals ) , rabbit monoclonal phosphoTyr705 STAT3 ( 1:30 , Cell signaling ) and mouse anti-CD163 ( 1:40 , Novus biologicals ) and rabbit anti-CD163 ( 1:40 , Novus Biologicals ) . Slides were washed in PBS and co-incubated with the appropriate AlexaFluor conjugated secondary antibodies ( 1:500 , anti-rabbit Alexa 488 and 1:500 anti-mouse Alexa 647 , Abcam ) in the dark for 1 hour . Slides were mounted with Vectashield DAPI mounting medium ( Vector Labs , Burlingame , CA ) to label DNA and imaged on a Zeiss Axioplan fluorescence microscope ( Oberkochen , Germany ) . Staining was quantified on ImageJ by counting the number of cells ( or DAPI nuclei for the transcription factors ) that stained positive for the indicated antibodies on at least three renal cortical fields at 20X magnification . Human renal proximal tubular epithelial cells immortalized with human papillomavirus 16 ( HK-2 cells ) were maintained in keratinocyte-serum free medium ( ATCC; Life Technologies , Carlsbad , CA ) supplemented with bovine pituitary extract , human recombinant epidermal growth factor and 1% penicillin-streptomycin ( Life Technologies ) . For all cytokine treatment experiments , cells were split into six-well plates and grown to 70% confluence . The cytokines used were recombinant human IL-6 ( Cell Signaling Technologies ) , recombinant human TNFα ( Cell Signaling Technologies ) and recombinant human IFNγ ( Cell Signaling Technologies ) . We used the drug S3I-201 ( Selleck Chemicals ) for STAT3 inhibition . For STAT3 inhibition experiments , cells received 50 uM of S3I-201 pre-treatment for 1 hour , followed by addition of 200 ng/mL human recombinant IL-6 for an additional 90 minutes . For TNFα and IFNγ stimulation experiments , cells were incubated with 100 ng/mL of recombinant TNFα or IFNγ respectively , or PBS vehicle control treatment . RNA was isolated at 90 minutes post-cytokine treatment using TRIZOL reagent ( Life Technologies ) according to the manufacturer’s protocol . Total RNA samples were submitted to the Stanford functional genomics DNA microarray core facility for assessment of RNA integrity . Each RNA sample was amplified using the Ambion Illumina RNA amplification kit with biotin UTP labeling . The Ambion Illumina RNA amplification kit uses T7 oligo ( dT ) primer to generate single stranded cDNA followed by a second strand synthesis to generate double-stranded cDNA , which was then column purified . In vitro transcription was performed to synthesize biotin-labeled cRNA using T7 RNA polymerase , and cRNA was column purified . The cRNA was then measured and total of 750 ng of cRNA was hybridized to array using standard Illumina protocols with streptavidin-Cy3 being used for signal detection . Slides were scanned on an Illumina Beadstation and analyzed using BeadStudio ( Illumina , Inc ) . Microarray expression data were normalized using the rank invariant normalization method [61] . Significant differential gene expression was determined using an unpaired Student’s t test for treated vs . control samples , and p-value thresholds used for statistical significance are indicated in each specific analysis , as appropriate . Only probe sets with detectable expression ( signal detection p < 0 . 01 in at least one experimental sample per batch of microarray experiments ) were analyzed for differential expression in subsequent analyses . For genes represented by multiple Illumina probe sets , the most differentially expressed probe set was used in subsequent analyses . Heat maps showing comparative gene expression patterns between different cytokine/transcription factor perturbations in HK-2 cells and in kidney aging were generated using MultiExperiment Viewer ( MeV version 4 . 0 ) software ( http://www . tm4 . org/ ) . Direct transcription factor targets for STAT1 , STAT3 and NFκB were defined as the subset of high-confidence ENCODE ChIP-seq targets from the proximal-filtered list published in [11] that also showed differential expression ( p < 0 . 05 in microarray experiments ) following the appropriate cytokine stimulation . For analysis of cytokine-regulated genes ( S3 Fig ) , we surveyed the entire transcriptome ( 9 , 508–10 , 121 detectably expressed probe sets ) for changes in gene expression , rather than restricting our analysis to the ChIP-seq binding targets for the three transcription factors ( 91 to 134 ChIP-seq binding targets ) . Therefore , we used a more stringent statistical threshold ( p < 0 . 001 ) for the identification of cytokine-regulated genes . Similarities between gene expression profiles were determined by the Pearson correlation between gene expression changes following inflammatory cytokine stimulation ( log2fold-change ) vs . gene expression changes during kidney aging ( age-related slope/beta coefficient ) ; the p-value for the concordance in expression pattern ( percentage of expression changes in a concordant direction ) between indicated gene sets was calculated using a binomial test . All microarray datasets generated in this study are publicly available in the Gene Expression Omnibus ( GEO Accession numbers: GSE68940 , GSE68941 , GSE68942 , GSE68826 ) . To assess the correlations between STAT1 , STAT3 and NFκB targets with macrophage markers , we analyzed published microarray data from [1] . STAT3 targets were defined as STAT3 ChIP-seq targets [11] that were also induced by IL-6 ( at a significance level of p < 0 . 05 ) from our microarray analysis . Similarly NFκB and STAT1 direct targets were defined as ChIP-seq targets that were induced by TNFα or IFNγ ( p < 0 . 05 ) and non-overlapping with the set of targets for the other two transcription factors . For each transcription target gene , we converted its normalized expression values into normalized z-scores across the 73 individual renal cortex samples and computed the averaged expression z-score . We then performed a linear regression analysis of average expression z-scores vs . chronological age to generate a linear regression equation for expression z-score vs . age . The averaged expression z-scores were converted to age-adjusted expression values based on the linear relationship between age and expression . Similarly , for transcripts specific to ( or strongly enriched in ) monocytes/macrophages ( e . g . CD163 , CD14 , TYROBP ) , each expression value was converted into a z-score , averaged and regressed on age to derive an age-adjusted expression level . Heat maps showing age-adjusted expression values for macrophage transcripts and transcription factor target genes were generated using MeV version 4 . 0 software ( Boston , MA ) . Rows and columns were hierarchically clustered by Pearson correlation . SNPs within the STAT1 , STAT3 and canonical NFκB transcription factor genes ( RELA and NFKB1 ) were selected using SCANdb ( http://www . scandb . org/newinterface/index . html ) and intersected with the list of all imputed SNPs from a previously published genome-wide association study of kidney function and chronic kidney disease [40 , 62] . In total , we tested 217 SNPs for their association with kidney function and chronic kidney disease . Thus , SNPs with an association p-value of less than 1 . 1 x 10−4 ( 0 . 05/434 tests ) remained statistically significant after Bonferroni correction for multiple testing . LocusZoom ( http://csg . sph . umich . edu/locuszoom/ ) was used to generate regional association plots for RELA and NFKB1 , illustrating the statistical association between each of the tested SNPs in RELA and NFKB1 with eGFR . | The structure and function of human kidneys deteriorate steadily with age , yet little is known about the underlying causes of kidney aging . In this work , we first used a genomics approach to identify candidate regulators of gene expression changes in the aging human kidney and identified inflammation-related transcription factors NFκB , STAT1 and STAT3 as the top candidate regulators . We found that kidney aging is associated with activation of NFκB , STAT1 and STAT3 in the renal parenchyma , and that the gene expression signatures evoked by activation of these transcription factors in human renal epithelial cells mimics age-associated gene expression changes in the kidney . Furthermore , we identified specific genetic variants in the NFκB transcription factor genes RELA and NFKB1 that associate with renal function and chronic kidney disease in humans , implicating NFκB as a potential contributor to the pathogenesis of chronic kidney disease and renal dysfunction in old age . Our findings suggest that activation of the inflammatory transcription factors STAT1 , STAT3 and NFκB underlie transcriptional changes and reduced renal function in the elderly . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | The Inflammatory Transcription Factors NFκB, STAT1 and STAT3 Drive Age-Associated Transcriptional Changes in the Human Kidney |
Campylobacter jejuni is one of the major causes of infectious diarrhea world-wide , although relatively little is know about its mechanisms of pathogenicity . This bacterium can gain entry into intestinal epithelial cells , which is thought to be important for its ability to persistently infect and cause disease . We found that C . jejuni is able to survive within intestinal epithelial cells . However , recovery of intracellular bacteria required pre-culturing under oxygen-limiting conditions , suggesting that C . jejuni undergoes significant physiological changes within the intracellular environment . We also found that in epithelial cells the C . jejuni–containing vacuole deviates from the canonical endocytic pathway immediately after a unique caveolae-dependent entry pathway , thus avoiding delivery into lysosomes . In contrast , in macrophages , C . jejuni is delivered to lysosomes and consequently is rapidly killed . Taken together , these studies indicate that C . jejuni has evolved specific adaptations to survive within host cells .
Campylobacter jejuni is the leading cause of bacterial food-borne illness in the United States and a major cause of diarrheal disease throughout the world [1] . C . jejuni infection is also an important pre-condition for Guillain-Barré paralysis [2] . Despite its public health importance , relatively little is known about its pathogenesis . Examination of intestinal biopsies of humans [3] , in vivo studies in infected primates [4] and other animal models [5–7] , together with in vitro experiments using cultured human intestinal epithelial cells [8–10] , have demonstrated that C . jejuni can invade non-phagocytic intestinal epithelial cells . However , to date , little is known about the molecular details of the mechanisms by which C . jejuni enters intestinal epithelial cells . Bacterial factors such as motility , glycosylation , and capsular synthesis have been implicated in C . jejuni internalization [11–14] . Strains with mutations in these pathways have defects in their ability to adhere to and invade host cells , as well as to colonize animals [12–19] . Bacterial invasion has also been correlated with C . jejuni's ability to stimulate the activation of MAP kinases leading to the production of the pro-inflammatory cytokine , IL-8 [20 , 21] . Taken together , these data suggest that bacterial internalization into intestinal epithelial cells is important in C . jejuni pathogenesis . Although most host factors that are required for C . jejuni internalization into non-phagocytic cells remain unknown , this entry process appears to have unique cytoskeletal requirements . Most other bacterial pathogens such as Listeria monocytogenes , Shigella flexneri , and Salmonella typhimurium utilize the host-cell actin cytoskeleton to gain intracellular access [22] . However , C . jejuni is internalized into intestinal epithelial cells in a microtubule-dependent , actin-independent fashion [10] , suggesting that this bacterium employs an entry mechanism unlike those reported for other bacterial pathogens . The intracellular fate of C . jejuni remains unknown , although it is likely that this bacterium , similar to other intracellular pathogens , may have evolved specific adaptations to survive within host cells . Intracellular pathogens utilize a variety of strategies to survive and replicate within host cells . For example , some pathogens such as Trypanasoma cruzi [23] , Listeria monocytogenes [24] , and Shigella flexneri [22 , 25] break out of the phagocytic vacuoles after internalization and can replicate within the cytosol of the infected cell . Other pathogens , such as Leishmania , have evolved an array of adaptations to survive in the hostile environment of the phagolysosome , which is characterized by low oxygen tension , poor nutrient content , low pH , and the presence of a variety of antibacterial products such as antibacterial peptides and lysosomal enzymes [26] . Yet another group of intracellular pathogens survive within a vesicular compartment that does not fuse with lysosomes . For example , Salmonella typhimurium [27] and Mycobacterium tuberculosis [28] alter the biogenesis and dynamics of their vacuolar compartment preventing fusion to lysosomes . All evidence to date indicates that after internalization into intestinal epithelial cells , C . jejuni resides within a membrane bound compartment [29–31] . We report here that C . jejuni survives within intestinal epithelial cells by deviating from the canonical endocytic pathway thus residing in a unique intracellular compartment that does not fuse with lysosomes .
Although C . jejuni internalization into host cells is believed to play a role in pathogenesis , little is known about its intracellular fate . We therefore examined the ability of C . jejuni to survive within intestinal epithelial cells . Human intestinal epithelial T84 cells were infected with C . jejuni , and total viable intracellular bacteria were determined at different times by counting colony forming units ( CFU ) . Significant numbers of CFU ( > 3 × 105/well ) were recovered at early time points , however , over time , the number of CFU recovered decreased considerably ( Figure 1A ) . By 24 h there was a ∼500-fold decrease in the number of CFU recovered from infected cells compared to 4 h after infection ( Figure 1A ) . These results suggest that intracellular C . jejuni rapidly loses viability during the course of its intracellular stage . This was surprising as it suggested that the ability of C . jejuni to enter non-phagocytic cells might not confer a significant advantage to this bacterium . We therefore examined the possibility that internalized C . jejuni may alter its physiology in such a way that , although viable , it may not be culturable under the conditions used in this assay . Indeed , C . jejuni has been reported to enter a viable but non-culturable state when subjected to a variety of stimuli or environments [32–34] . To address this issue , we stained C . jejuni recovered from cultured intestinal T84 cells with reagents that distinguish viable from non-viable bacteria ( see Materials and Methods ) . Using these reagents , we observed no decrease in viability of intracellular C . jejuni over time ( Figure 1B and 1C ) . In fact , FACS analysis also revealed that the ratio of viable to non-viable bacteria did not change over the course of infection ( Figure 1C and 1D ) . These results indicate that C . jejuni remains viable for at least 24 h after infection and suggest that it acquires a physiological state that does not allow the recovery of CFU under our standard culture conditions . We hypothesized that once internalized by intestinal epithelial cells , C . jejuni might adapt to the low oxygen environment within the cell by changing its mode of respiration . We therefore tested whether the intracellular bacterial population could be cultured if allowed to “recover” under conditions in which oxygen is very limiting . Human intestinal epithelial T84 cells were infected with C . jejuni and the number of CFU was determined after culturing under oxygen-limiting incubation or under 10% CO2 conditions . The number of CFU decreased drastically ( ∼500 fold ) when bacteria were directly grown under an atmosphere of 10% CO2 ( Figure 2A ) or in GasPak jars ( BBL Microbiology Systems , Cockeysville , MD ) with BBL CampyPacks ( BBL Microbiology Systems , Cockeysville , MD ) ( data not shown ) . However , there was no significant decrease in the number of CFU recovered over time when the plates were incubated under oxygen-limiting conditions for 48 h and then switched to an atmosphere of 10% CO2 ( Figure 2A ) . The number of CFU recovered when plates where pre-incubated under low-oxygen conditions closely correlated with the number of viable bacteria quantified via FACS analysis ( Figure 2B ) . These data demonstrate that C . jejuni remains viable within intestinal epithelial cells for at least 24 h , although it undergoes physiological changes such that requires exposure to oxygen-limiting conditions for its efficient recovery . There are contradictory reports regarding the ability of C . jejuni to survive within macrophages [29 , 35–38] . Given our findings in intestinal epithelial cells suggesting physiological changes in intracellular C . jejuni , which affects its ability to be cultured , we re-examined its ability to survive within macrophages . Mouse primary bone marrow–derived macrophages ( BMDM ) were infected with C . jejuni and the number of CFU recovered over time was determined by incubating culture plates directly under either 10% CO2 or subjecting them to a 48 h pre-incubation under oxygen-limiting conditions prior to incubation under 10% CO2 . We found a significant decrease in the number of CFU recovered over time regardless of the culture condition . By 12 h there was a severe reduction in the number of CFU recovered under both 10% CO2 or with pre-incubation under oxygen-limiting conditions ( >2 , 000 and >100 fold , respectively , compared to the CFU recovered 1 h after infection ) ( Figure 2C ) , and no CFU were recovered at 24 h of infection regardless of the culture conditions . Taken together , these data demonstrate that intracellular C . jejuni can survive within intestinal epithelial cells but are killed by professional phagocytes . Immediately after internalization into host cells , C . jejuni resides within a membrane bound compartment [29–31] . It is therefore possible that in order to survive within cells , this bacterium has evolved specific adaptations to survive within lysosomes or to modulate host cellular trafficking events to avoid fusion with lysosomes . If C . jejuni survives within lysosomes , the C . jejuni–containing vacuole ( CCV ) should be accessible to endocytic tracers . To test this hypothesis , Cos-1 cells were first infected with C . jejuni and subsequently exposed to the fluorescent endocytic tracer dextran , which was chased into lysosomes . As a control , cells were infected with a strain of S . typhimurium carrying a mutation in invA and a plasmid expressing the Yersinia pseudotuberculosis protein invasin . InvA is an essential component of the invasion-associated type III secretion system [39] and therefore this strain enters cells through the invasin-mediated pathway [40] . Subsequent to its uptake , the invasin-expressing bacteria is delivered to lysosomes ( Watson and Galán , unpublished data ) . Immunofluorescense microscopy analysis revealed that 90% of the vacuoles containing S . typhimurium invA ( invasin ) colocalized with the endocytic tracer dextran ( Figure 3A and Figure S1 ) . In contrast , only ∼15% of the C . jejuni–containing vacuoles acquired detectible amounts of dextran ( Figure 3A and Figure S1 ) . These data suggest that the CCV is functionally separated from the canonical endocytic pathway . To confirm these results , we used another endocytic tracer , gold-labeled bovine serum albumin ( BSA-gold ) [41] . Cos-1 cells were first infected with C . jejuni or S . typhimurium invA ( invasin ) and subsequently exposed to BSA-gold , which was chased into lysosomes ( see Materials and Methods ) . BSA-gold was then imaged using electron microscopy to determine if the CCV was accessible to this fluid phase endocytic tracer . Although BSA-gold colocalized with ∼75% of the vacuoles containing S . typhimurium invA ( invasin ) ( Figure 3B , 3C and 3I ) , only 15% of the CCVs were accessible to the endocytic marker ( Figure 3D , 3E and 3I ) . Furthermore , the CCVs that colocalized with BSA-gold appeared to be morphologically different from the CCVs that did not . The CCVs accessible to the endocytic tracer were spacious and contained additional electron dense materials , closely resembling lysosomes ( Figure 3F ) . In contrast , the CCVs that did not colocalize with BSA gold had tight membranes around the bacteria and the compartments did not resemble lysosomes ( Figure 3D and 3E ) . These data confirm the results obtained using fluorescence microscopy and provide additional evidence that the CCV is segregated from the canonical endocytic pathway . Since our experiments established that C . jejuni quickly loses viability within macrophages , we hypothesized that in these cells this bacterium may be delivered to lysosomes . To test this hypothesis , BMDM were infected with C . jejuni and then incubated with media containing BSA-gold and examined by electron microscopy as described in Materials and Methods . In contrast to what we observed in epithelial cells , in macrophages over 90% of the CCVs were readily accessible to the endocytic tracer ( Figure 3G , 3H and 3I ) . Similar results were obtained using fluorescent dextran as an endocytic tracer ( data not shown ) . Taken together , these data indicate that in macrophages , C . jejuni is delivered to lysosomes where it cannot survive , while in intestinal epithelial cells C . jejuni is segregated from an endocytic pathway leading to lysosomes and consequently survives in a vacuolar compartment that is distinct from lysosomes . We next examined whether avoidance of lysosomal delivery was essential for C . jejuni survival within epithelial cells . To this end , we carried out an experiment in which C . jejuni was internalized via the Fc receptor , a pathway known to lead to lysosomes [42] . Cos-1 cells expressing the murine Fc receptor were infected with either opsonized or non-opsonized C . jejuni and at different times after infection the CFU were determined by plating under the permissive oxygen-limiting conditions and further incubation in 10% CO2 environment . As shown in Figure 4A , the relative number of CFU recovered from cells infected with opsonized C . jejuni 24 h after infection was significantly ( > 20 fold ) lower than the number of CFU recovered from cells infected with non-opsonized bacteria . These results indicate that internalization via the Fc receptor results in a significant loss of intracellular viability , presumably because these bacteria are ultimately delivered to lysosomes . To confirm this hypothesis , we examined whether C . jejuni internalized via the Fc receptor was accessible to an endocytic tracer . Cos-1 cells expressing the mouse Fc receptor were infected with opsonized C . jejuni and subsequently exposed to fluorescent dextran . Consistent with the hypothesis that the loss of viability of C . jejuni internalized via the Fc receptor was due to its delivery to lysosomes , >80% of the opsonized bacteria colocalized with fluorescent dextran , compared to ∼20% of non-opsonized control ( Figure 4B ) . Taken together , these data show that C . jejuni is unable to survive within lysosomes and further indicate that this bacterium has evolved a mechanism to avoid delivery to this compartment in order to survive within intestinal epithelial cells . Furthermore , these results also indicate that the mechanism of bacterial entry into host cells has a major impact in the ability of C . jejuni to survive intracellularly . To investigate the biogenesis and trafficking of the CCV , we examined the dynamics of acquisition of both early and late endosomal markers . Cos-1 cells were infected as described in Materials and Methods and at different times after infection the presence of different endocytic markers was probed by immunofluorescence microscopy using specific antibodies . Fifteen minutes after infection the majority ( ∼65% ) of the intracellular bacteria co-localized with the early endosomal marker EEA-1 ( Figure 5A and Figure S2 ) . However , by 60 min , only ∼20% of the CCV co-localized with EEA-1 and more than 80% were stained by an antibody directed to the late endosomal marker lamp-1 ( Figure 5B and Figure S2 ) . Two hours after infection , almost all CCVs stained with lamp-1 . The acquisition of lamp-1 may therefore occur via an alternative pathway since , as shown above , the CCV does not fuse with lysosomes . In an effort to better understand the nature of the C . jejuni compartment , we tested the CCV for the presence of other markers of the early and late endocytic pathway . We found that early in infection , 65–70% of the CCVs co-localized with the early endosomal markers Rab4 , Rab5 , and with a probe for phosphoinisitide 3 phosphate ( green fluorescent protein fused to the PX domain of the 40 kD subunit of the nicotinamide adenine nucleotide phosphate oxidase ) ( Figure 5B-5D , Figure S3 and Video S1 ) . Two hours after infection , less than 10% of the CCVs colocalized with any of these markers indicating that the CCV interacts transiently with these compartments ( Figure 5B-5D ) . Furthermore , C . jejuni transiently acquires the late endosomal marker Rab7 ( Figure 5E , 5F and Figure S4 ) . At 45 min after infection , ∼65% of CCVs acquired Rab7-GFP , while at 2 h , only ∼20% of CCVs colocalized with Rab7-GFP . However , consistent with the observation that the mature CCV does not co-localize with endocytic tracers , the lysosomal marker cathepsin B was seen in only a very small proportion of the CCVs , even at 2 h after infection ( Figure 5H and Figure S4 ) , although it was present in 95% of vacuoles containing S . tyhimurium invA ( invasin ) ( Figure 5G and Figure S4 ) . These data further indicate that C . jejuni survives within a unique intracellular compartment that despite harboring the lamp-1 protein , is functionally distinct from lysosomes . Our results indicate that at some point after internalization , the CCV deviates from the canonical endocytic pathway . Therefore , we set out to determine at what stage of the endocytic pathway this segregation might occur . The GTPases Rab5 and Rab7 are involved in the biogenesis of early and late endosomes , respectively [43] . Overexpression of dominant negative forms of these GTPases disrupt temporal and spatial delivery of internalized cargo to lysosomes [43] . To determine if acquisition of lamp-1 by the CCV required Rab5 or Rab7 , Cos-1 cells were transfected with wild type or dominant negative forms of Rab5 ( Rab5S34N ) or Rab7 ( Rab7N125I ) . The transfected cells were infected with C . jejuni and the acquisiton of lamp-1 by the CCV was assessed by immunofluorescence microscopy . As a control , a similar experiment was conducted using S . typhimurium invA ( invasin ) , which traffics to lysosomes . Overexpression of Rab5S34N and Rab7N125I did not affect lamp-1 acquisition by the CCV , although it effectively prevented acquisition of this marker by the vacuoles containing S . typhimurium invA ( invasin ) ( Figure 6 and Figure S5 ) . These results demonstrate that the CCV acquires lamp-1 by an alternative pathway apparently segregating from the canonical endocytic pathway early after C . jejuni internalization . Collectively , our data suggest that the mechanism by which C . jejuni enters epithelial cells may ultimately determine its intracellular fate . Although internalization through Fc receptors delivers C . jejuni to lysosomes , when entering via its own specific adaptations C . jejuni segregates from the endocytic pathway and avoids delivery to lysosomes . The mechanisms of C . jejuni internalization are unusual in that they do not require the actin cytoskeleton and are dependent on microtubules [10] . In fact , disruption of the actin cytoskeleton increases the efficiency of bacterial uptake ( Figure S6 ) . Previous studies have shown that addition of filipin , an agent that sequesters cholesterol , decreased the ability of C . jejuni to enter into cultured epithelial cells [44 , 45] , suggesting that lipid rafts or caveolae may be required for efficient entry into cells . Consistent with this observation , we found that addition of the cholesterol-sequestering agent methyl-beta cyclodextrin ( MβCD ) blocked C . jejuni internalization into T84 in a dose-dependent manner ( Figure S6 ) . Similar results were obtained with Cos-1 cells ( data not shown ) . To further investigate the potential involvement of lipid rafts or caveolae in C . jejuni internalization , we examined the CCV for the acquisition of caveolin-1 and flotillin-1 , two markers associated with these membrane domains [46–49] . Cos-1 cells were transfected with plasmids encoding GFP-tagged forms of caveolin-1 or flotillin-1 , and the association of these markers with the CCV was examined by time lapse and fluorescence microscopy as described in Materials and Methods . C . jejuni acquired caveolin-1-GFP and flotillin-1-GFP immediately after internalization ( Figure 7A and 7B ) . Quantification of this association determined that at early time points during infection , ∼60% of the CCVs colocalized with both caveolin-1-GFP ( Figure 7C ) and flotillin-1-GFP ( Figure 7D and Video S2 ) . The association , however , was transient since at later points after infection <10% of the CCVs were seen in association with these markers ( Figure 7C and 7D ) . Vesicles devoid of bacteria but labeled by flotillin-1-GFP were also observed immediately after C . jejuni internalization , and some of them eventually fused with the nascent CCV ( Video S3 and S4 ) . Caveolin-1 and flotillin-1 have been shown to be involved in various endocytic events , including the internalization of microbial pathogens [50] . We therefore further examined the potential involvement of caveolin-1 or flotillin-1 in C . jejuni internalization into non-phagocytic cells . Depletion of flotillin-1 by siRNA did not result in a measurable decrease in the abiliy of C . jejuni to enter cells ( p = 0 . 13 ) ( Figure S7 ) . However , expression of a dominant interfering mutant of caveolin-1 ( caveolin-1Y14F ) significantly decreased C . jejuni internalization ( p = 0 . 02 ) ( Figure 7E ) . Taken together these data indicate that caveolin-stabilized lipid membrane domains ( i . e . , caveolae ) are important for C . jejuni efficient entry into non-phagocytic cells . The GTPase dynamin is involved in pinching off of the nascent endosome in both clathrin- and caveolae-mediated endocytosis [51–54] . We therefore tested the potential involvement of dynamin II in C . jejuni internalization . Cos-1 cells were transfected with a plasmid expressing a dominant negative form of dynamin II ( dynIIK44A ) , which has been shown to inhibit both clathrin and caveolae-dependent endocytosis [52–54] , and the ability of C . jejuni to enter those cells was examined by fluorescence microscopy as described in Materials and Methods . Although expression of dynIIK44A-GFP effectively blocked the uptake of transferrin ( data not shown ) , it did not inhibit the uptake of C . jejuni ( Figure 7F ) . In fact , there was a modest enhancement of C . jejuni internalization in the presence of wild-type dynamin II . These results indicate that the role of caveolin-1 C . jejuni entry into cells may not be related to its role in caveolae-mediated endoyctosis . Rather , caveolin-1 or caveolae may play a role in the signaling events leading to bacterial uptake , perhaps by facilitating the spatial organization of critical signaling molecules . In fact , tyrosine kinases have been reported to be essential for C . jejuni entry into host cells [44 , 45 , 55] , a result that we have confirmed ( Figure S6 ) . Since efficient signaling through receptor tyrosine kinases is known to require lipid rafts or caveolae , it is possible that the inhibitory effect of cholesterol sequestering agents or dominant negative caveolin-1 may be the result of interference with tyrosine kinase signaling . Examination of the localization of the CCV over time showed that at 4–5 h after infection , C . jejuni localized to the perinuclear region [56] . To gain more insight into the specific localization of the CCV in relation to other organelles , we investigated by immunoflurescence microscopy the position of the CCV in relation to the Golgi apparatus over time using an antibody directed against GM130 , a Golgi resident protein . Two hours after infection , intracellular C . jejuni were evenly distributed around the cell ( Figure 8A ) . However by 6–8 h of infection , >85% of the CCVs were seen in close association with the Golgi apparatus ( Figure 8B and 8E ) , close to the microtubule organizing center ( Figure S8 ) . The close association of the CCV and the Golgi does not represent a default pathway for any internalized particle traveling to a perinuclear position since internalized S . typhimurium invA ( invasin ) did not show association to the Golgi despite the fact that these phagosomes were also located in a perinuclear region ( Figure 8C ) . Electron microscopy analysis confirmed the intimate association of the CCV and the Golgi apparatus ( Figure 8D ) . However , the CCV did not acquire Golgi markers ( data not shown ) , indicating that despite their close association , the two compartments remain distinct . We then investigated the mechanism by which the CCV reaches its perinuclear destination . Many intracellular bacteria in membrane-bound compartments traffic along microtubule tracks to reach their destination within the cell . We first investigated the role of microtubules in the localization of the CCV . When nocodazole was added to disrupt the microtubule network after infection of Cos-1 cells , the CCVs were observed distributed throughout the cell and did not reach a perinuclear location ( Figure 8F ) . These data show that intact microtubules are necessary for the CCV to reach its final destination . Phagosomes most often move along microtubules through the action of specific motors [57 , 58] . Cytoplasmic dynein is a minus-end directed motor that is responsible for moving cargo away from the periphery and toward the microtubule organizing center [59] and is therefore a candidate motor to move the CCV to its final destination . We investigated this hypothesis by overexpressing GFP-dynamatin p50 , a subunit of the dynactin complex that when overexpressed , blocks the function of dynein [60] . Immunofluorescence analysis showed that overexpression of dynamatin p50 effectively disrupts the localization of C . jejuni at 6 h post-infection ( Figure 8G ) . These results are consistent with a previous observation indicating that addition of orthovanadate , a rather non-specific inhibitor of dynein , inhibits the movement of the CCV to a perinuclear position [56] . Taken together , these results indicate that subsequent to internalization , the CCV travels to a perinuclear position in the immediate vicinity of the Golgi apparatus and that the movement of the CCV requires both microtubules and the molecular motor dynein .
Similar to other enteric pathogens , C . jejuni has evolved the ability to gain intracellular access to non-phagocytic intestinal epithelial cells and this process has been implicated in pathogenesis [7 , 14 , 20 , 61 , 62] . Although most work to date has focused on C . jejuni entry into host cells , the intracellular fate of this pathogen has been largely uncharacterized . We have shown here that C . jejuni survives within intestinal epithelial cells , although over time it acquires a metabolic state that renders it unculturable under standard culture conditions . However , C . jejuni recovered from within epithelial cells could be cultured if subjected to conditions of severe oxygen limitation . These results suggest that once within epithelial cells , C . jejuni may either become oxygen sensitive or may alter its respiration mode so that it can no longer be cultured in the presence of oxygen . The recently completed nucleotide sequence of the genome of the C . jejuni strain 81–176 used in this study revealed the presence of genes involved in additional respiratory pathways , including electron acceptors that may be utilized for alternate modes of respiration [63] . Thus , these additional respiration genes may contribute to the ability of C . jejuni 81–176 to survive within intestinal epithelial cells . We showed here that C . jejuni survives within intestinal epithelial cells within a compartment that is distinct from lysosomes ( Figure 9 ) . CCVs are not accessible to endocytic tracers indicating that they are functionally separated from described endocytic pathways leading to lysosomes . In fact , when targeted into lysosomes after internalization via the Fc receptor , C . jejuni was unable to survive within epithelial cells . These results indicate that C . jejuni has evolved specific adaptations to traffic within host cells and avoid delivery into lysosomes . These adaptations may be important to faciliate colonization of the host by providing a safe-heaven where C . jejuni can avoid innate immune defense mechanisms . However , those adaptations must not be able to operate in macrophages since , in these cells , C . jejuni is targeted to lysosomes and therefore cannot survive . Our results suggest that C . jejuni deviates from the canonical endocytic pathway shortly after internalization ( Figure 9 ) . The CCV appears to interact with early endosomal compartments since it associates with early endosomal markers such as EEA-1 , Rab5 , Rab4 , and PX-GFP ( which labels PI3P ) . However , this interaction is transient and does not lead to progression within the canonical endocytic pathway . The presence of markers of lipid-associated rafts and caveolae on the CCV suggests that C . jejuni may reside in a compartment that is functionally distinct from early endosomes . In fact , C . jejuni was still able to target properly in the presence of dominant interfering mutants of Rab5 or Rab7 , which control early events in the endocytic pathway [43] . C . jejuni has unusual cytoskeletal requirements to gain intracellular access to intestinal epithelial cells since its internalization is dependent on microtubules but not on the actin cytoskeleton [10] , as is usually the case for most intracellular bacteria [22] . Consistent with previous observations suggesting that caveolae are required for C . jejuni entry [44 , 45] , we have shown here that bacterial internalization is dependent on caveolin-1 . However , we showed that the entry process is independent of dynamin , whose function is essential for clathrin and caveolae-mediated endocytosis [51–53] . In fact , expression of a dominant-inhibitory form of dynamin resulted in a reproducible increase in the ability of C . jejuni to enter cells . We therefore hypothesize that a caveolin-1-stabilized lipid membrane domain may be required for proper signaling through tyrosine kinases , which are also required for C . jejuni internalization rather than for endocytosis . In fact , efficient signaling through receptor tyrosine kinases requires lipid rafts or caveolae [64 , 65] and the surface availability of many these is regulated by dynamin [66] . In this context , we hypothesize that the enhancement of C . jejuni internalization observed when inhibiting dynamin function , may be the result of an increase in the availability of putative surface “receptor” for this pathogen resulting in enhanced signaling for entry . What is the nature of the CCV ? We showed that the CCV contains lamp-1 , although this compartment is unique and clearly distinct from lysosomes since it does not colocalize with the lysosomal protein marker cathepsin B and it is not accessible to endocytic tracers . In fact , the acquisition of lamp-1 , which occurs very early in the CCV maturation , must occur by an unusual mechanism that does not require the GTPases Rab5 or Rab7 . Interestingly , S . typhimurium resides within a vacuole that is apparently segregated from the canonical endocytic pathway [27 , 67] and also harbors lamp-1 , although in this case acquisition of this marker appears to require Rab7 [68] . Another unique property of the CCV is its close association with the Golgi , which requires microtubules and the motor protein dynein . More studies will be required to better define the nature of this compartment and the precise mechanisms by which C . jejuni modulates vesicular trafficking . In summary , we have established that C . jejuni has evolved specific adaptations to survive within intestinal epithelial cells by avoiding delivery into lysosomes . This survival strategy does not appear to operate in BMDM since C . jejuni is rapidly killed in these cells . We hypothesize that C . jejuni's unusual entry mechanism may be central to its ability to avoid delivery into lysosomes since when internalized via a different pathway ( e . g . , via the Fc receptor ) , C . jejuni could not avoid delivery into lysosomes . Its diversion from a pathway leading to lysosomes must therefore occur upon entry . Understanding the mechanism by which this bacterium survives within host cells may provide new insights into C . jejuni pathogenesis as well as reveal undiscovered paradigms in host cellular trafficking .
Wild-type C . jejuni 81–176 has been described previously [69] . C . jejuni were routinely grown on tryptic soy broth agar supplemented with 5% sheep blood ( BA ) or in brain heart infusion ( BHI ) broth at 37 °C under 10% CO2 , or where indicated , in an anaerobic chamber under low oxygen conditions ( BD-Diagnostic Systems GasPak Plus Anaerobic System Envelopes with Palladium Catalyst , catalog number 271040 , New Jersey ) , or with BBL CampyPacks ( BBL Microbiology Systems , Cockeysville , MD ) . S . typhimurium invA has been described previously [70] and was transformed with invasin-encoding plasmid pRI203 , which mediates mammalian cell entry via αβ1 integrins [71] . A S . typhimurium invA strain expressing the dsRed protein was constructed as follows . The plasmid DsRed . T3_S4T , which expresses the dsRed protein under the control of an arabinose-inducible promoter [72] , was digested with EcoRI and ScaI to release a fragment containing dsRed and the paraABC promoter . This fragment was ligated into pACYC184 and resulting plasmid , pSB3082 , was transformed into S . typhimurium invA ( pRI203 ) . S . typhimurium invA expressing invasin and dsRed was routinely grown in LB containing tetracycline ( 10 μg ml−1 ) , ampicillin ( 50μg ml−1 ) , and 0 . 1% arabinose to induce DsRed expression . T84 , a human intestinal epithelial cell line , and Cos-1 , a monkey kidney epithelial cell line , were obtained from the American Type Culture Collection ( Rockville , MD ) and were grown in DMEM supplemented with 10% fetal bovine serum containing penicillin ( 100 U ml−1 ) and streptomycin ( 50 μg ml−1 ) . Bone marrow–derived macrophages ( BMDM ) , were obtained as previously described [73] . Briefly , femurs and tibias were excised and flushed with DMEM containing 10% fetal bovine serum ( FBS ) , penicillin ( 100 U ml−1 ) , and streptomycin ( 50 μg ml−1 ) . Cells were spun down and resuspended in BMDM differentiation medium [DMEM containing 20% FBS , 30% L-cell supernatant , penicillin ( 100 U ml−1 ) and streptomycin ( 50 μg ml−1 ) ] and plated onto non-tissue culture treated 10-cm2 plastic dishes . The cells were fed fresh BMDM differentiation medium on day 3–4 to allow further differentiation until day 6–7 . BMDM were then seeded in the appropriate tissue culture dishes to be used in infection experiments . C . jejuni was harvested from a fresh BA plate and grown in BHI broth under under 10% CO2 until mid log phase ( OD600 = 0 . 7–0 . 8 ) . To prepare the inoculum , bacteria were pelleted at 20 , 000 × g in a microfuge for 2 min and directly resuspended in Hank's Balanced Salt Solution ( HBSS ) ( Invitrogen ) . The inoculum was diluted in HBSS to adjust for different multiplicity of infections ( MOI ) . Serial dilutions of the inoculum were plated onto BA plates to determine the number of bacteria . T84 cells were split to 70% confluence ( ∼105 cells per well ) in a 24 well dish . BMDM were seeded at 2 × 105 cells per well in a 24 wells dish . After washing 3X with HBSS , T84 cells and macrophages were infected with an MOI of 50 or 20 , respectively . The plates were centrifuged at 200 × g for 5 min to maximize bacteria-cell contact and incubated for 1 or 2 h at 37 °C 5% CO2 . Following the incubation , the monolayers were washed 3X with HBSS and incubated with complete media containing gentamicin ( 100 μg ml−l ) for 2 h . This concentration of gentamicin was found to be optimal to kill extracellular bacteria without affecting the viability of intracellular bacteria . Plating of the infection medium determined that no significant number of c . f . u . were detected after this treatment . For experiments involving longer time points , the media was replaced with complete media containing gentamicin ( 10 μg ml−1 ) . Again , plating of the infection medium determined that no significant number of c . f . u . were present after this treatment . After 3 additional washes , the infected cells were lysed at the designated time points and the samples were prepared for colony forming unit ( CFU ) determination or FACS analysis . To quantify the number of intracellular bacteria , cells were lysed in PBS with 0 . 1% deoxycholate and the CFU were enumerated after plating serial dilutions grown at 37°C with 10% CO2 , or in GasPak jars ( BBL Microbiology Systems , Cockeysville , MD ) with BBL CampyPacks ( BBL Microbiology Systems , Cockeysville , MD ) . For anaerobic incubations , plates were incubated in an anaerobic chamber with a GasPak ( BD-Diagnostic Systems GasPak Plus-Anaerobic System Envelopes with Palladium Catalyst , Catalog number 271040 , New Jersey ) for 48 h and incubated further at 37 °C with 10% CO2 . For immunofluorescence studies , Cos-1 cells were seeded on glass cover slips in 24 well plates . C . jejuni was cultured as described above . S . typhimurium invA expressing invasin was cultured in LB containing ampicillin ( 30 μg ml−1 ) overnight and subcultured 1:20 for 3 h prior to infection . For time courses , cells were infected with an MOI of 100 and 50 for C . jejuni and S . typhimurium invA ( invasin ) , respectively , centrifuged from 5 min at 1 , 000 × g to maximize bacteria-host cell contact , and incubated for an additional 15 min at 37 °C 5% CO2 . Wells were washed three times in PBS and either fixed in 4% PFA for a 15 min time point or the media was replaced with DMEM + 10% FBS with gentamicin ( 100 μg ml−1 ) to kill the extracellular bacteria and prevent additional bacterial internalization . At later time points cells were washed an additional three times in PBS , and fixed in 4% paraformaldehyde . After 4 h or longer infection times , no significant number of bacteria that had remained extracellular but attached to the cell were detected using the gentamicin treatment described ( i . e . , initial addition of 100 μg/ml for 2 h and subsequent addtion of 10μg/ml for the remainder of the experiment , see above ) ( Figure S9 ) . For quantitation of intracellular bacteria in transfected cells , Cos-1 cells were infected with an MOI of 25 for 1 h as described above . Cells were washed and incubated for 2 additional hours in DMEM + 10% FBS with gentamicin ( 100 μg ml−1 ) to kill the extracellular bacteria and prevent additional bacterial internalization . The cells were then fixed and processed for immunofluorescence as described below . Where indicated , cells were incubated with Nocodazole ( 10 μM ) , Cytocholasin D ( 5 μM ) , Genistein ( 10 μM and 100 μM ) , dissolved in DMSO ( Sigma ) , or methyl-beta cyclodextrin ( MβCD ) dissolved in PBS ( 1mM-10mM ) 30 min prior to infection and kept throughout the duration of the incubation period . All control cells were treated with the appropriate solvent for the same length of time . T84 cells and BMDM from wild-type mice were seeded at density of 105 cells per well on a 24-well dish and infected with an MOI of 50 and 20 , respectively . Following a 1 h incubation at 37 °C and 5% CO2 , the cells were washed with HBSS and DMEM containing 10% FBS and 100 μg ml−1 gentamicin was added to each well . Cells were washed again and lysed at the designated time points in 500 μl of 0 . 05% sodium deoxycholate in PBS . The cell lysates were collected and subjected to a low speed spin ( 1 , 000 rpm ) for 2 min to remove large cell debris . Supernatants were collected and intracellular bacteria were isolated by a 2 min high-speed spin ( 10 , 000 rpm ) . The isolated bacterial pellet was resuspended in 500 μl filter-sterilized staining buffer ( PBS containing 1mM EDTA and 0 . 01% Tween ) . The bacteria were then stained with the reagents of a cell viability kit ( BD Biosciences , San Jose , CA ) , which distinguishes live and dead cells by using a thiazole orange ( TO ) solution , which stains all bacteria , and propidium iodide ( PI ) , which only stains dead bacteria . TO and PI were added to final concentrations 53 nM and 11 μM , respectively , in accordance with the manufacturer's instructions . After 5 min of staining , bacteria were pelleted , washed once in PBS , resuspended in 1 ml of PBS and analyzed by flow cytometry . The absolute count of live/dead bacteria was carried out by addition of 50 μl of a liquid suspension of a known number of fluorescent beads ( supplied in the kit , BD Biosciences , San Jose , CA ) following the manufacturer's instructions . Samples were analyzed on a FACS calibur flow cytometer . TO fluoresces primarily in FL1 and FL2; PI primarily in FL3 . An SSC threshold was used , and cells and beads were gated using scatter and FL2 , which detects the TO fluorescence and therefore the total bacterial population . In order to best discriminate between live and dead bacteria , an FL1 versus FL3 plot was used and live and dead populations were gated within this plot ( dead cells , FL3+; live cells , FL1+ ) . To determine the bacterial concentration , the following equation was used: # events in cell region/ # events in bead region x # beads/test/test volume × dilution factor = concentration of cell population . A plot was generated after using this equation to calculate the number of viable bacteria ( in triplicate wells ) in both T84 and BMDM at each time point . Cos-1 cells were infected as described above and at the designated time points , cells were washed three times in PBS and fixed in 4% paraformaldehyde ( PFA ) for 13 min at room temperature ( RT ) . The fixed cells were washed three times in PBS and permeabilized by incubating them in PBS containing 3% non-fat milk and 0 . 05% saponin ( PBS-MS ) ( Calbiochem ) . Cover slips were incubated in primary antibody diluted in PBS-MS for 30 min . The cover slips were then washed three times in PBS and incubated in secondary antibody . After two washes in PBS and two washes in deionized water , the cover slips were mounted onto glass slides using Prolong Gold antifade reagent ( Molecular Probes ) . Images were acquired on a Nikon TGE2000-U Eclipse inverted microscope fitted with a Micromax Princeton digital camera controlled by the Metamorph software package , version 6 . 1 ( Universal Imaging Corp . , Downingtown , PA ) . When needed , inside-out staining was used to differentiate extracellular from intracellular bacteria . Briefly , before permeabilization with saponin , extracellular bacteria were stained with rabbit anti-C . jejuni antiserum in PBS containing 3% milk followed by Alexaflour 350-conjugated anti-rabbit antibodies ( Molecular Probes ) . Cells were washed three times , permeabilized , and the total bacterial population was stained with rabbit anti-C . jejuni followed by Alexaflour 594-conjugated mouse anti-rabbit antibodies ( Molecular Probes ) . After two washes in PBS and two washes in deionized water , the cover slips were mounted onto glass slides using Prolong Gold antifade reagent ( Molecular Probes ) . Where indicated , nuclei stained with DAPI ( Invitrogen ) . Rabbit antibodies against C . jejuni were obtained by repeated immunization of rabbits with a mixture of equal amounts of formaldehyde and heat-killed whole cell C . jejuni . Anti-S . typhimurium ( rabbit ) antibodies were purchased from DIFCO Laboratories , Detroit , Michigan . Mouse monoclonal antibodies against EEA-1 , Lamp-1 , beta-tubulin , and GM130 were acquired from BD Biosciences Pharmingen . A mouse anti cathepsin B antibody was a gift from the laboratory of Dr . Ira Mellman , Yale University , New Haven , CT . Secondary antibodies used were: Alexa 596- Alexa-488 , Alexa 350-conjugated goat anti-rabbit and Alexa 596- Alexa 488- Alexa 350-conjugated goat anti-mouse IgG antiserum ( Molecular Probes ) . Eukaryotic expression vectors encoding GFP-tagged wild-type Rab5 and Rab7 as well as their dominant negative mutants have been previously described [74–76] . GFP-tagged Rab4 , PX , dynamin II , dynamin IIK44A , caveolin-1 , and caveolin-1Y14F as well as murine Fc-receptorII ( FcRII ) expressing eukaryotic vectors have been described elsewhere [77–81] . The human flotillin-1 gene was amplified from a human cDNAs library by polymerase chain reaction ( PCR ) using primers fwd ( 5′-TAGCTCGAGCCATGTTTTTCACTTGTGGCCC-3′ ) and rev ( 5′-TCTAGAATTCCGGCTGTTCTCAAAGGCTTGA-3′ ) . The PCR product was then cloned into pEGFP-N1 ( Clontech , Oxford , UK ) using XhoI and EcoR1 . The resulting plasmid , pSB3111 , yielded flotillin-1 fused to the N-terminus of GFP . pFlot1-FLAG was a generous gift from Dr . Rosanna Paciucci , Unitat de Recerca Biomedica , Barcelona , Spain [82] . Plasmid DNA was purified using the Maxiprep kit ( Qiagen ) and used for transfection of cells with LipofectAMINE 2000 ( Invitrogen ) according to the manufacturers instructions . To quantitate the percentage of CCV containing different endocytic fluid tracers or cellular markers , infected cells were visualized directly in the fluorescence microscope . Using the Metamorph software package a series of images were taken , including internalized bacteria , total bacteria , and the cellular marker . Overlayed fluorescent images were analyzed by determining the number of CCVs that contained the corresponding marker . A minimum of one hundred vacuoles was analyzed per cover slip for each treatment and designated post-infection time . Each experiment was completed in triplicate wells/cover slips and expressed as an average . CCVs were considered positive for the presence of a marker when they contained detectable amounts of the staining probe/antibody . The same methodology was used to quantitate the number of intracellular bacteria within transfected cells . Briefly , Cos-1 cells seeded on cover slips were transfected with plasmids encoding GFP-tagged proteins , infected , and processed for immunofluorescence . A minimum of 50 transfected cells were randomly selected and imaged as described above and the number of intracellular bacteria was quantitated within each cell . Each experiment was completed in triplicate wells/cover slips and expressed as an average number of bacteria per transfected cell and where designated normalized to control cells . Statistical analysis of the results was carried out by the student t-test . Cos-1 cells were grown on coverslips in 24-well dishes and infected with C . jejuni or S . typhimurium invA ( invasin ) and dsRed at an MOI of 25 and 10 respectively . After spinning for 5 min at 200 × g and allowing the infection to proceed for 1 h , the cells were washed three times and fresh media containing gentamicin was added to kill the extracellular bacteria . After an additional 3 h the cells were incubated with Texas Red or Alexafluor-488 labeled dextran ( 1 mg/ml; Molecular Probes ) for 1 h . The cells were washed three times and incubated with fresh media containing gentamicin for an additional 2 h . The cover slips were then processed for immunofluorescence . Cos-1 cells were plated in a 24 well dish at ∼40% confluency and transfected with a plasmid encoding FcRII [80] . After an overnight incubation , cells were incubated with C . jejuni that were opsonized with rabbit polyclonal anti-C . jejuni antibodies ( 1:1 , 000 ) . Plates were centrifuged at 200 × g for 5 min and incubated at 37 °C 5% CO2 for 1 h , after which the cells were washed three times with HBSS , and incubated with complete media containing gentamicin ( 100 μg ml−1 ) . Cells were then lysed at the designated time points and the number of viable intracellular bacteria was assessed as described above . Cos-1 cells were grown on life cell imaging dishes ( MatTek Corp . ) and transfected with flotillin-1-GFP or PX-GFP [78] encoded eukaryotic expression vectors as described above . C . jejuni was cultured as described above and fluorescently labeled using PKH26 Red Flourescent Cell Linker Kit according to manufacturer's instructions ( Sigma ) . The labeling protocol did not affect C . jejuni's viability or its ability to enter into non-phagocytic cells ( data not shown ) . Cells were infected with an MOI of ∼100 . Time-lapse series were acquired using a Nikon TGE2000-U Eclipse inverted microscope fitted with a Micromax Princeton digital camera controlled by the Metamorph software package , version 6 . 1 ( Universal Imaging Corp . , Downingtown , PA ) . Filters allowed the simultaneous detection of GFP and rhodamnine , respectively . The acquired images were merged as RGBs ( for two color movies ) and converted into QuickTime movies using Metamorph . For the flotillin-1-GFP three dimensional movie , cells transfected with flotillin-1-GFP were infected for 15 min , fixed , and processed for imunoflourescence using inside-out staining . Z-stacks were acquired using a Zeiss Axio Imager Upright fluorescent Microscope fitted with an Apotome and the AxioCam MRc5 digital camera controlled by the Axiovision software package , version 4 . 2 ( Carl Ziess MicroImaging , Inc . ) . The cropped images were reconstructed and converted into a 3D QuickTime movie using Axiovision . Depletion of flotillin-1 was performed using a pool of three target-specific 20–25 nucleotide siRNAs designed to knock down gene expression ( Santa Cruz Biotechnology , Inc . , Santa Cruz , CA ) . The siRNA pool was transfected into Cos-1 cells using LipfectAMINE 2000 ( Invitrogen ) . To mark transfected cells , flotillin siRNA was cotransfected with pEGFP-N1 at a ratio of 5:1 . Intracellular bacteria within GFP-expressing cells were quantified as described above . RNAi silencing efficiency and specificity were analyzed at the protein level by Western blot analysis 72 h after cotransfection of pFlot1-FLAG [82] with the siRNA pool . Protein depletion by RNAi was normalized to endogenous levels of actin using rabbit anti-actin antibodies ( Sigma-Aldrich ) . Cos-1 cells or BMDM were plated at approximately 8 × 106 cells on 10 cm2 plastic tissue culture dishes . After 3x wash with warm HBSS , C . jejuni , cultured as described above , were used to infect cells at an MOI of 100 . At 1 . 5 h post-infection , the extracellular bacteria were washed off three times with warmed HBSS and the plates were incubated with warmed culture media containing gentamicin to kill the extracellular bacteria . For standard EM , the infected cells were incubated at 37 °C 5% CO2 for and additional 4 . 5 h . For BSA-gold experiments , after a 3 . 5 h incubation , the infected cells were incubated for 1 h with BSA-gold-containing complete media . After three additional washes cells were fixed in situ with a freshly made solution of 1% glutaraldehyde ( from an 8% stock from Electron Microscopy Sciences ( EMS ) , Fort Washington , PA ) 1% OsO4 in 0 . 05 M phosphate buffer at pH 6 . 2 for 45 min . After fixation , the cells in petri plates were rinsed three times with cold distilled water and en bloc stained with uranyl acetate overnight . The petri plates were then dehydrated in ethanol then placed into hydroxypropyl methacrylate ( EMS ) , which does not react with the plastic in the petri dish , and embedded in L 112 , an epon substitute ( Ladd , Burlington , VT ) . Following polymerization of the epon , the block was cut out and mounted and thin sections were cut through their exposed surfaces . Thin sections were collected on naked grids stained with uranyl acetate and lead citrate and examined in a Philips 200 electron microscope . At least fifty vacuoles were analyzed for the presence of BSA-gold for each condition . | Campylobacter jejuni is one of the most common causes of food-borne illness in the United States and a major cause of diarrheal disease throughout the world . After infection through the oral route , this bacterium invades the cells of the intestinal epithelium , a property that is important for its ability to cause disease . Usually , bacteria and other material entering the cell move to compartments called lysosomes , where an acidic mix of enzymes breaks it down . This study shows that C . jejuni can survive within intestinal epithelial cells by avoiding delivery to lysosomes . In contrast , in macrophages , which are specialized cells with the capacity to engulf and kill bacteria , C . jejuni cannot avoid delivery into lysosomes and consequently is rapidly killed . These studies help explain an important virulence attribute of C . jejuni . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology",
"infectious",
"diseases",
"microbiology",
"eubacteria"
] | 2008 | Campylobacter jejuni Survives within Epithelial Cells by Avoiding Delivery to Lysosomes |
Rainfall patterns are one of the main drivers of dengue transmission as mosquitoes require standing water to reproduce . However , excess rainfall can be disruptive to the Aedes reproductive cycle by “flushing out” aquatic stages from breeding sites . We developed models to predict the occurrence of such “flushing” events from rainfall data and to evaluate the effect of flushing on dengue outbreak risk in Singapore between 2000 and 2016 . We used machine learning and regression models to predict days with “flushing” in the dataset based on entomological and corresponding rainfall observations collected in Singapore . We used a distributed lag nonlinear logistic regression model to estimate the association between the number of flushing events per week and the risk of a dengue outbreak . Days with flushing were identified through the developed logistic regression model based on entomological data ( test set accuracy = 92% ) . Predictions were based upon the aggregate number of thresholds indicating unusually rainy conditions over multiple weeks . We observed a statistically significant reduction in dengue outbreak risk one to six weeks after flushing events occurred . For weeks with five or more flushing events , compared with weeks with no flushing events , the risk of a dengue outbreak in the subsequent weeks was reduced by 16% to 70% . We have developed a high accuracy predictive model associating temporal rainfall patterns with flushing conditions . Using predicted flushing events , we have demonstrated a statistically significant reduction in dengue outbreak risk following flushing , with the time lag well aligned with time of mosquito development from larvae and infection transmission . Vector control programs should consider the effects of hydrological conditions in endemic areas on dengue transmission .
Dengue , a disease transmitted by Aedes mosquitoes , is a global public health problem . Dengue is endemic in more than 100 tropical and subtropical countries where 30–50% of the global population is at risk for infection [1–4] . Annually , there are an estimated 50–100 million cases of dengue , with 500 , 000 of these cases developing into life-threatening Dengue Hemorrhagic Fever and Dengue Shock Syndrome [5] . Various weather factors influence dengue incidence . Temperature and humidity impact dengue incidence by affecting adult feeding behavior , larvae development , and mosquito survival [2 , 5–16] . Small increases in average monthly temperature ( e . g . , 1°C ) have been associated with a considerable increase in dengue incidence , leading to a 45% increase in the number of cases in subsequent months in Brazil and China [4 , 5 , 17–30] . Although mosquitoes require sufficient rainfall for breeding and larval development [2 , 7–9 , 31 , 32] , too much rainfall can be detrimental [33–35] . Excessive rainfall can cause breeding sites to overflow , disrupting mosquito breeding and destroying developing larvae . Mosquito breeding site “flushing” , where water levels exceed a breeding site’s drainage threshold and wash away mosquito larvae , has been observed in both experimental and field settings [33–36] . In experimental studies , simulated heavy rainfall washed away the majority of mosquito larvae and resulted in significant larvae mortality [33 , 35 , 36] . The extent of the effect of flushing depended on rainfall intensity , container size , and larvae age [33 , 35] . A field study in Singapore demonstrated that dengue incidence is lowest following months where flushing events are most frequent suggesting that flushing events may influence how and when dengue transmission occurs [34] . Multiple studies have described the weather drivers of dengue incidence using regression and time series models [4 , 5 , 10 , 17–30 , 37–39] . While relationships between temperature and humidity and dengue incidence have been consistent , the relationship between rainfall and dengue incidence has remained unclear . Associations between rainfall and dengue incidence have ranged from weak or no connection [19 , 25 , 27–29 , 39] , to as much as a 21% increase in dengue incidence in response to increased rainfall [10 , 19 , 26 , 37] . Machine learning tools have been used to predict the occurrence of dengue based on a combination of weather parameters , including rainfall [40–48] . Though successful in predicting weekly and monthly dengue incidence with over 90% accuracy , these models did not explain the relationship between rainfall and dengue . One interesting and still unanswered question relates to the influence of flushing on dengue incidence . Observations by Seidahmed and Eltahir [34] of larvae survival in storm drains in Singapore following rainfall events have confirmed that flushing mostly occurs during the Northeast monsoon season . The authors noted that the season with the highest rainfall levels and most flushing events preceded the season with traditionally low dengue incidence . However , the study did not quantify the rainfall patterns leading to flushing nor its effect on dengue spread . This paper proposes a quantitative approach associating flushing , expressed through rainfall patterns , with the subsequent fluctuations in dengue outbreak risk in Singapore between 2000 and 2016 . The methodology uses entomological data from Singapore [34] combined with historical rainfall data to quantify the rainfall conditions associated with mosquito larvae wash-out , or “flushing” . A regression model is then used to estimate the effect of flushing events on dengue outbreaks . We show a statistically significant 16–70% reduction in dengue outbreak risk in one to six weeks following flushing events .
Singapore is located on the southern-most tip of the Malay Peninsula with a population of 5 . 6-million people [49] . Dengue is hyper-endemic , where serotypes I-IV co-circulate , and is transmitted year-round , with peak incidence occurring between July and September [50–53] . Singapore has a tropical rainforest climate ( Köppen: Af ) with two monsoon seasons , the Northeast and Southwest monsoons . The former is associated with heavy rainfall between November and March while the latter occurs between June and October and is relatively drier [54] . Average annual precipitation is nearly 2 . 3 meters . Average daily temperature is stable throughout the year , where average daily temperatures of the hottest and coolest months differ by 1 . 9°C . Weekly dengue case counts in Singapore from 2000–2016 ( N = 887 ) were obtained from the Weekly Infectious Disease Bulletin of the Singapore Ministry of Health [55] . Confirmed cases were reported by all public and private hospitals and laboratories who are mandated to report all clinically and lab diagnosed cases of dengue within 24 hours [56 , 57] . Longitudinal entomological surveys were obtained from a previously published work of Seidahmed and Eltahir [34] . These surveys were conducted in the Geylang neighborhood of Singapore , a highly urbanized neighborhood located east of the Singapore River [34] . Geylang is classified as being hyperendemic for dengue by the National Environmental Agency since dengue transmission occurs year-round [34] . Geylang is estimated to have a resident population of 32 , 000 and an even larger non-resident population which is a result of the large amount of cheap housing options that are primarily used by foreign laborers [34] . During entomological data collection , random aquatic surveys were completed twice a week between August 2014 and August 2015 ( except between February 21st and March 10th ) , resulting in 107 days of entomological observations . For each survey , trained inspectors examined all potential outdoor natural and artificial mosquito breeding sites ( e . g . , open and closed roadside storm drains and non-drain sites such as canvas sheets , pails and flowerpots ) , looking for mosquito aquatic stages in randomly selected neighborhood blocks . Samples of aquatic stages were taken and evaluated for taxonomic classification . A subsample of aquatic specimens was retained until adult emergence to confirm taxonomic identification . Taxonomic keys [58–60] were used to classify sampled aquatic stages and emerged adults [34] . A total of 6 , 824 samples were taken from potential breeding sites ( 5 , 818 samples from open and closed storm drains ) [34] . Sixty-seven breeding sites ( 53 occurring in storm drains ) were positive for Ae . aegypti breeding [34] . Particular attention was then given to the 53 positive Ae . aegypti breeding sites that were found in open and closed storm drains [34] . Breeding sites of Ae . Aegypti were mainly found in the southern part of Geylang where denser urban drainage network and low-rise housing predominate [61] . The 53 positive breeding sites occurring in storm drains were continuously monitored for hydrological conditions and changes in the presence of mosquito larvae . For each visit during the monitoring phase , the following four conditions were observed: 1 ) stagnant water and positive for aquatic stages , 2 ) stagnant water and negative for aquatic stages , 3 ) dry and negative for aquatic stages , and 4 ) flushed and negative for aquatic stages [34] . No sites were identified as flushed and positive for aquatic stages or dry and positive for aquatic stages . We then classified each day of observation as “Flushed” if at least one breeding site was classified as “flushed and negative” , meaning that water had exceeded the drainage threshold ( indicated by the storm drain overflowing , or evidence of an overflow ) for the breeding site and mosquito larvae were not present . Of the 107 days of entomological observations , 25 were classified as flushed and 82 as non-flushed ( Fig 1 ) . The majority of flushing events ( 84% ) occurred during the Northeast monsoon , while only 23% of non-flushing events happened during this time . This data was used to develop a model to predict flushing occurrence for the entire study period . Daily weather data , including rainfall and temperature , were obtained from Tanjong Katong weather station ( selected for its proximity to Geylang ) [62] . Humidity data were extracted from remote sensed reanalysis data [63] and El Niño Southern Oscillation ( ENSO ) data were obtained from the Climate Prediction Center [64] , both of which are operated by the National Oceanic and Atmospheric Administration . The ENSO index used was the normalized Oceanic Niño Index for Niño region 3 . 4 [65–68] and used to define the ENSO phase ( i . e . , El Niño , La Niña , Neutral ) . Weather data were obtained from 1/1/1999–12/31/2016 . Missing weather data were imputed using multiple imputations through chained equations using the MICE R package [69] . Entomological data detailing the hydrological conditions of breeding sites in Singapore were only available for a single year ( August 2014 –August 2015 ) during the study period ( 2000–2016 ) . We developed the Predictive fLUshing-Mosquito model ( PLUM ) which predicts the occurrence of flushing events based on the temporal variation in daily rainfall over several weeks preceding the day of interest and allows the extension of the prediction to dates when no entomological observations are available . The objective was not only to make a daily prediction but to identify more general ‘flushing’ conditions leading to drains’ overflow . Early flooding warning systems use rainfall thresholds to predict flooding occurrence [70] . The PLUM model operates in a similar fashion by identifying a set of variables and their thresholds associated with flushing occurrence . The PLUM model was developed using the entomological observations and the corresponding rainfall data between August 2014 to August 2015 . The general model framework can be found in S1 Fig . The proposed approach identifies rainfall thresholds associated with a higher likelihood of flushing occurrence . We used supervised machine learning models to associate the identified thresholds with flushing occurrence . Each model was trained using a balanced training set where non-flushed observations were randomly under-sampled to generate a 1:1 ratio of flushed to non-flushed observations in the training set to prevent the model from classifying all observations as the majority class [71] . Each model was evaluated on unseen data using leave-one-out cross validation . The PLUM model was then extended to the entire study period , to predict the occurrence of flushing events in Singapore between 2000 and 2016 .
We applied UFA to identify thresholds associated flushing . Running UFA for all 38 rainfall variables , we identified 36 thresholds ( 20 for cumulative rainfall variables , 16 for daily rainfall variables ) associated with an increased likelihood of flushing occurrence . The likelihood of flushing occurrence was found to increase when variable values were greater than or equal to the associated high risk thresholds . All UFA identified thresholds can be found in S1 Table . In total , there were 107 days of entomological observations , 25 were defined as flushed and 82 were defined as not flushed . We used the PLUM model ( described in section 2 . 5 ) to classify each entomological observation using leave-one-out cross validation . The PLUM model made predictions based upon two variables , the aggregate number of high risk thresholds that were met per day for both cumulative and daily rainfall variables . The PLUM model achieved 92% accuracy and demonstrated a strong ability to discriminate between flushed and non-flushed observations ( Table 2 ) . There is a well-defined threshold resulting in nearly perfect separation between flushed and non-flushed observations ( Fig 2 ) . We fit Eq ( 1 ) to each cross validation sample and identified the mean value for each model coefficient resulting in the following model: FlushedD=0 . 56 ( ThresholdsMetD , Daily−HighRisk ) +0 . 20 ( ThresholdsMetD , Cumulative−HighRIsk ) +7 . 94 . ( 5 ) We observed that the fit of Eq ( 1 ) for each cross validation sample was stable indicating good generalizability of the PLUM model . We extended the PLUM model to the entire study period . Using Eq ( 5 ) we assigned a value indicating whether or not flushing occurred on each day within the study period . For the study period , 1 , 242 ( 21 . 2% ) days were classified as flushing , a similar proportion to the number of flushed days ( 25 , 23 . 4% ) in the observed entomological data time period . These results were used in creating the “weekly flushing” variable which is an aggregator of daily flushing events per week . We used a distributed lag nonlinear logistic regression model to evaluate the association between flushing occurrence and the risk of a dengue outbreak in the weeks following the flushing events . Dengue incidence was reported in 887 weeks between the years 2000 and 2016 . Summary statistics for outbreak occurrence , flushing occurrence , and other weather variables during the study period are presented in Table 3 . During the study period , 138 ( 15 . 6% ) weeks were defined as an Outbreak week based upon the selected criteria . There is also evidence of seasonal variation in outbreak occurrence and rainfall . Outbreak weeks were at least three times as likely to occur during the Southwest monsoon compared with the Northeast and Non-monsoonal periods . Total weekly rainfall shows a seasonal pattern where weeks during the Northeast monsoon have on average 15-20mm more rainfall than other weeks . Due to increased rainfall , flushing events were most likely to occur during the Northeast monsoon as compared with any other season . Fig 3 shows the prevalence of flushing events and the prevalence of outbreak weeks by month . Here we observe a negative association in which months where flushing event prevalence is highest ( November to February ) the outbreak week prevalence is low . Moreover , in months where flushing event prevalence is low ( June to September ) , outbreak week prevalence is highest . Regression analysis , Eq ( 2 ) , demonstrates a negative association between flushing events and dengue outbreak risk ( Fig 4 ) . We identified a nonlinear association between the number of flushing events per week and dengue outbreak risk that varied over the lag dimension . The risk of an outbreak occurring was significantly lower for weeks where five or more flushing events occurred compared with weeks with zero flushing events; this relationship remained significant up to six weeks after the flushing events occurred ( Table 4 ) . Weeks where seven flushing events occurred , there was between a 30–70% reduction in the risk of an outbreak up to six weeks after the flushing events occurred . Smaller reductions in dengue outbreak risk were also observed when five ( 16–38% reduction in risk up to four weeks after the flushing events ) and six ( 24–56% reduction in risk up to five weeks after the flushing events ) flushing events occurred in a week .
We proposed a non-linear approach to understanding the relationship between excessive rainfall , flushing , and dengue outbreak occurrence in Singapore . According to the PLUM model , flushing conditions are characterized by rainfall patterns indicating excess rainfall . We demonstrated that rainfall-induced flushing is associated with a statistically significant decreased risk of dengue outbreak , with association being significant up to six weeks after the week when flushing occurred . | Dengue transmission is sensitive to fluctuations in rainfall and other weather conditions because it is transmitted by the Aedes mosquito . Recent studies have identified that extreme rainfall can result in mosquito breeding site flushing . However , these rainfall conditions have neither been described nor evaluated for their potential effect on dengue transmission . In this study , we applied machine learning and regression approaches to identify rainfall thresholds associated with mosquito breeding site flushing in Singapore . We then estimated the association between the number of flushing events per week and dengue outbreaks in the following weeks . Here we demonstrate that flushing events are accurately predicted by historical rainfall patterns . We also show that flushing events are associated with a statistically significant reduction in dengue outbreak risk up to six weeks after the flushing events occurred . This research suggests that dengue predictive and early warning systems must consider hydrological conditions and other contributing factors to accurately predict near-term dengue risk . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"monsoons",
"atmospheric",
"science",
"geographical",
"locations",
"animals",
"singapore",
"seasons",
"developmental",
"biology",
"storms",
"insect",
"vectors",
"zoology",
"infectious",
"diseases",
"life",
"cycles"... | 2018 | Statistical modeling of the effect of rainfall flushing on dengue transmission in Singapore |
Disruption of certain genes can reveal cryptic genetic variants that do not typically show phenotypic effects . Because this phenomenon , which is referred to as ‘phenotypic capacitance’ , is a potential source of trait variation and disease risk , it is important to understand how it arises at the genetic and molecular levels . Here , we use a cryptic colony morphology trait that segregates in a yeast cross to explore the mechanisms underlying phenotypic capacitance . We find that the colony trait is expressed when a mutation in IRA2 , a negative regulator of the Ras pathway , co-occurs with specific combinations of cryptic variants in six genes . Four of these genes encode transcription factors that act downstream of the Ras pathway , indicating that the phenotype involves genetically complex changes in the transcriptional regulation of Ras targets . We provide evidence that the IRA2 mutation reveals the phenotypic effects of the cryptic variants by disrupting the transcriptional silencing of one or more genes that contribute to the trait . Supporting this role for the IRA2 mutation , deletion of SFL1 , a repressor that acts downstream of the Ras pathway , also reveals the phenotype , largely due to the same cryptic variants that were detected in the IRA2 mutant cross . Our results illustrate how higher-order genetic interactions among mutations and cryptic variants can result in phenotypic capacitance in specific genetic backgrounds , and suggests these interactions might reflect genetically complex changes in gene expression that are usually suppressed by negative regulation .
Cryptic genetic variants are standing polymorphisms that only exhibit phenotypic effects under atypical conditions , such as when specific genes are compromised or the environment dramatically changes [1–3] . Work in Arabidopsis thaliana ( e . g . , [4–6] ) , Caenorhabditis elegans ( e . g . , [7–9] ) , Drosophila melanogaster ( e . g . , [10–14] ) , multiple budding yeasts ( e . g . , [15–19] ) , and a number of non-model organisms ( e . g . , [20–26] ) has shown that cryptic variation is abundant within and between species . Because it is so prevalent , cryptic variation could plausibly contribute to adaptation and phenotypic novelty [2 , 27–29] , as well as to disease susceptibility [30] . Yet due to their entirely conditional phenotypic effects , cryptic variants have proven difficult to study and are not understood as well as other classes of polymorphisms . In particular , the genetic and molecular mechanisms that suppress and uncover cryptic variation have yet to be fully determined . For the purposes of this paper , we focus on the mechanisms by which functional disruption of specific ‘capacitor’ genes exposes the phenotypic effects of cryptic variants . This phenomenon is often referred to as ‘phenotypic capacitance’ or ‘evolutionary capacitance’ , though for simplicity we refer to it as ‘capacitance’ [11 , 31] . The first described capacitor was Hsp90 , a chaperone that assists in the folding and stabilization of other proteins [11 , 32] . Early research on capacitance suggested that Hsp90 might have distinct biochemical features that cause cryptic variation to be uncovered when it is compromised [4 , 11 , 32] . However , subsequent theoretical work showed that capacitance most likely occurs as a general consequence of gene regulatory network perturbation and that many genes might be able to act as capacitors [31] . Supporting this finding , a number of genes involved in chromatin regulation have also been shown to be capacitors of cryptic variation [15 , 33 , 34] and to even phenocopy the effects of Hsp90 perturbation [34] . More recent work suggests that capacitance depends not only on the perturbation of capacitors but also on the specific cryptic variants that are present . This is because cryptic variants themselves can play an important role in capacitance by genetically interacting with and ‘potentiating’ the phenotypic effects of their capacitors [3 , 17 , 33 , 35–37] . The genetic architecture of this potentiating cryptic variation has not been characterized in detail [38] , but may involve complex epistatic interactions between multiple cryptic variants and capacitating mutations ( i . e . , higher-order genetic interactions ) [39] . In such a scenario , the phenotypic effect of a given capacitating mutation would depend on the cryptic variants with which it co-occurs , with the mutation having an effect only in certain genetic backgrounds [40] ( Fig 1 ) . This possibility is not unfounded , as several recent studies suggest that genetic background effects can involve higher-order genetic interactions among de novo or induced mutations and sets of cryptic variants [41–43] . We recently described an experimental system that can be used to study how higher-order genetic interactions among mutations and cryptic variants result in capacitance [42] . In our previous paper , we showed that a de novo mutation in IRA2 , a negative regulator of the Ras-cAMP-PKA ( Ras ) pathway [44 , 45] , uncovers sets of interacting cryptic variants that influence colony morphology in Saccharomyces cerevisiae . This mutation ( ira2Δ2933 ) occurred spontaneously while we were generating a cross of the lab strain BY4716 ( ‘BY’ ) and a derivative of the clinical isolate 322134S ( ‘3S’ ) [46 , 47] , and results in a truncated , partially functional Ira2 protein that lacks 117 amino acids relative to its wild type form . When the ira2Δ2933 lesion is present in specific haploid recombinants in the BYx3S cross , it causes a change in colony morphology from ‘smooth’ to ‘rough’ ( Fig 2 ) . Through comprehensive genetic mapping experiments , we showed that ira2Δ2933 induces the rough phenotype when it co-occurs with specific combinations of cryptic variants at four or more genes [42] . To better understand these higher-order genetic interactions , we cloned all of the genes involved in one of the combinations . This resulted in the identification of two transcriptional activators that heterodimerize and function downstream of the Ras pathway ( FLO8 [48] and MSS11 [49] ) , a structural protein that plays a role in vesicle formation ( END3 [50 , 51] ) , and an enzyme that helps cells detoxify themselves of endogenous redox stress ( TRR1 [52] ) . Most of the rough individuals in our past study had the genotype END3BY FLO83S ira2Δ2933 MSS11BY TRR13S . However , we also provided evidence for a more complex genotype involving END33S that requires specific alleles at two additional loci . In this paper , we complete our efforts to determine the genetic basis of ira2Δ2933-dependent rough morphology in the BYx3S cross under our standard assay conditions . We show that in addition to the previously identified five-way genetic interaction , a six-way interaction can also cause the trait . Specifically , individuals with the genotype END33S FLO83S ira2Δ2933 MSS11BY exhibit the rough phenotype if they possess BY alleles at two other transcription factors that are regulated by the Ras pathway [53 , 54]: the activator MGA1 [55] and the repressor SFL1 [56 , 57] . This suggests that the rough phenotype arises due to genetically complex changes in the regulation of Ras target genes . We examine the role of ira2Δ2933 in these regulatory changes and find that it alleviates the silencing of FLO11 , a gene that encodes a cell surface protein required for rough morphology . We also show that this ability to disrupt FLO11 repression is not unique to IRA2 . These results illustrate how higher-order combinations of cryptic variants can confer the potential for capacitance to specific genetic backgrounds and indicate that capacitating mutations may reveal cryptic phenotypic potential by causing transcriptional derepression .
To determine the specific combination of alleles involved in rough morphology in an END33S background , we generated new mapping populations by mating an END33S rough segregant from a ( BYx3S ) x3S backcross to BY and 3S ( Methods ) . Throughout the paper , the term ‘backcross’ refers specifically to these ( ( BYx3S ) x3S ) xBY and ( ( BYx3S ) x3S ) x3S matings . Because END33S segregated in the BY backcross , we genotyped rough individuals recovered from this population to determine the allele of END3 they carried ( Methods ) . In total , we obtained 63 and 88 rough END33S individuals from the BY and 3S backcrosses , respectively . We then pooled cells from these rough individuals by cross and performed bulk segregant mapping by sequencing [58 , 59] ( Methods ) . We found that the more complex genetic interaction involves a specific combination of alleles at six loci , with individual loci detected on Chromosomes V , VII , XIII , and XIV , and two loci identified on Chromosome XV ( Fig 3A and 3B ) . The chromosome XIV locus corresponds to END33S , while allele replacements in a backcross segregant that carried the six-way interaction confirmed that FLO83S , MSS11BY , and ira2Δ2933 underlie the Chromosome V , XIII , and XV-1 loci , respectively ( Fig 3C and S1 Fig; Methods ) . The new mapping data also allowed us to delimit the Chromosome VII and XV-2 loci , which we were unable to clone in our prior study [42] , to a single gene ( MGA1 ) and five genes ( SFL1 , ARP8 , LSC1 , SUF5 , THI80 ) , respectively . We used allele swaps to show that the BY alleles of MGA1 and SFL1 , which respectively encode an activator and a repressor that are regulated by the Ras pathway , are the causal alleles at these loci ( S1 Fig ) . These results show the six-way interaction occurs in individuals with the genotype END33S FLO83S ira2Δ2933 MGA1BY MSS11BY SFL1BY ( Fig 3B ) . Thus , the differences between the five- and six-way interactions involve which END3 allele is involved and whether specific alleles of MGA1 , SFL1 , and TRR1 are required ( Fig 3B ) . Based on our genetic mapping results in this paper and our past work [42] , we have identified alleles of six genes ( END3 , FLO8 , MGA1 , MSS11 , SFL1 , TRR1 ) that genetically interact in two different combinations with ira2Δ2933 ( Fig 3B and 3C ) . We tested whether these two allele combinations fully explain rough morphology in the BYx3S ira2Δ2933 cross by generating a new BYx3S cross in which 3S carried ira2Δ2933 ( Methods ) . As our past work focused on matings of segregants to BY or 3S , this population enabled us to test for the first time the effects of all possible combinations of BY and 3S alleles in the presence of ira2Δ2933 . Among 42 rough individuals that we recovered , 40 ( 95 . 2% ) carried the five-way interaction , while two ( 4 . 8% ) carried the six-way interaction . The five-way interaction should occur twice as often as the six-way interaction , yet the observed ratio was 20:1 . This may be due to linkage between END3 and a locus at which the BY allele confers a strong selective advantage during random spore isolation ( see Figure S2B from [42] ) . Alternatively , the enrichment of rough individuals carrying the five-way interaction could simply have occurred because the sample of rough individuals in this experiment was small . Nevertheless , our observation that all the examined rough individuals harbored either the five- or six-way interactions suggests that we have completely determined the genetic basis of rough morphology in the BYx3S ira2Δ2933 cross under our experimental conditions . Rough morphology in the BYx3S cross likely arises due to genetically complex changes in the regulation of Ras target genes . Such a possibility is supported by the finding that four Ras-regulated transcription factors [54] harbor cryptic variants involved in the rough phenotype , as well as by the fact that these cryptic variants are revealed by a capacitating mutation in IRA2 , a negative regulator of Ras signaling . A gene that is likely influenced by these genetic factors is FLO11 , which encodes a cell surface glycoprotein that facilitates cell-cell adhesion and is thought to be regulated by Flo8-Mss11 , Mga1 , and Sfl1 [60 , 61] . To determine if expression of the rough phenotype due to the five- and six-way interactions requires FLO11 , we deleted the gene from a nearly isogenic line possessing the five-way interaction and a backcross segregant carrying the six-way interaction ( Methods ) . This was sufficient to convert both of these strains from rough to smooth ( Fig 4A ) , indicating that both genetic interactions are FLO11-dependent . RT-PCR showed that FLO11 is expressed in individuals carrying the five- and six-way interactions , but not in BY or 3S ( Fig 4B; Methods ) . These results suggest expression of the rough phenotype requires active transcription of FLO11 . We tested whether ira2Δ2933 influences FLO11 expression by introducing the lesion into BY and 3S , and conducting RT-PCR ( Methods ) . Each strain remained smooth after this manipulation , which was expected because they both lack a complete set of alleles that can give rise to rough morphology . Furthermore , BY ira2Δ2933 did not express FLO11 , likely because this strain carries a nonsense allele of FLO8 , the major transcriptional activator of FLO11 [62] . However , introduction of ira2Δ2933 into 3S , which possesses a functional allele of FLO8 , converted FLO11 from a silenced to an actively transcribed state ( Fig 4B ) . Given that ira2Δ2933 alleviated repression of FLO11 in 3S , we hypothesized that it might do so by indirectly inhibiting Sfl1 , which is thought to negatively regulate FLO11 and other targets of the Ras pathway when Ras signaling is low by recruiting the Ssn6-Tup1 corepressor complex [57] , which in turn recruits the histone deacetylase Hda1 [63 , 64] . To test this possibility , we deleted SFL1 from 3S . This knockout phenocopied the results of introducing ira2Δ2933: 3S remained smooth , but expressed FLO11 ( Fig 4B ) . This suggests that iraΔ2933 disrupts Sfl1-mediated transcriptional repression of Ras target genes . To test whether loss of transcriptional repression by Sfl1 is sufficient to reveal the cryptic higher-order genetic interactions that specify rough morphology , we generated new BYx3S crosses . We first created a BYx3S cross that lacked the IRA2 mutation and screened for rough morphology among thousands of recombinants ( Methods ) . All segregants in this cross were smooth . We then constructed a cross in which BY and 3S carried wild type alleles of IRA2 , but had SFL1 deleted ( Methods ) . Rough morphology , as well as a ‘bumpy’ intermediate phenotype that we previously reported ( see Figure S4D and S1 Table in [42] , as well as S1 Note ) , segregated in this sfl1Δ cross ( Fig 5A ) . Genotyping of 44 rough sfl1Δ segregants showed that the rough phenotype is expressed in the ira2Δ2933 and sfl1Δ backgrounds largely due to the same cryptic variants ( Methods ) . 43 ( 98% ) of the rough sfl1Δ segregants possessed the genotype END3BY FLO83S MSS11BY TRR13S , which also potentiates the five-way interaction involving ira2Δ2933 ( Fig 5B ) . The other rough sfl1Δ segregant had the genotype END3BY FLO83S MSS11BY TRR1BY , which does not give rise to rough morphology in the presence of ira2Δ2933 ( Fig 5B ) . None of the rough sfl1Δ segregants had a genotype resembling the six-way interaction involving ira2Δ2933 . This could have occurred because SFL1BY , which is required for the six-way interaction , is missing from the sfl1Δ cross; our sampling was biased due to the selectively advantageous locus that is linked to END3; or , as the detection of a rough sfl1Δ segregant with the END3BY FLO83S MSS11BY TRR1BY genotype also suggests , ira2Δ2933 and sfl1Δ have similar but not identical molecular effects . Despite these differences between the ira2Δ2933 and sfl1Δ crosses , our results clearly show that transcriptional repression normally suppresses rough morphology and that multiple genes can act as capacitors by disrupting this negative regulation .
Across this manuscript and our previous paper [42] , we have cloned six genes that harbor cryptic variants that interact in two specific allele combinations to determine the phenotypic effect of ira2Δ2933 . These two genetic backgrounds can be viewed as potentiating genotypes that facilitate the expression of rough morphology in the presence of a capacitating mutation , such as ira2Δ2933 . This finding is important because it shows sets of cryptic variants can genetically interact with each other and their capacitating mutation , and implies a conceptual link between capacitance , higher-order genetic interactions , and genetic background effects ( Fig 1 ) . Given that four of the identified genes encode transcription factors , our work suggests complex gene regulatory changes underlie the expression of rough morphology in the BYx3S cross . This finding is consistent with theoretical results that have shown an important role for gene regulatory network perturbation in capacitance [31] and higher-order genetic interactions [65] . In our specific case , the role of ira2Δ2933 is likely to cause transcriptional derepression , which may enable the involved cryptic variants to collectively alter the gene regulatory network underlying colony morphology . Supporting such a role for derepression in the rough phenotype , we have shown that IRA2 is not unique in its ability to act as a capacitor . Rather , SFL1 can also serve as a capacitor of rough morphology , presumably because its deletion also causes transcriptional derepression . Moving forward , fully understanding capacitance in the BYx3S colony morphology system will likely require defining the gene regulatory network underlying rough morphology and determining how it changes across combinations of cryptic variants and capacitating mutations . Such work can shed light on the individual and collective contributions of the identified cryptic variants to the rough phenotype; may reveal why MGA1BY , SFL1BY , and TRR13S only have phenotypic effects in specific END3 backgrounds; and might further clarify how multiple genes can act as capacitors of the same cryptic variants and trait . More generally , research along these lines has the potential to provide basic insights into how genetically complex , cryptic phenotypes are suppressed and uncovered . Additionally , to our knowledge , the present study , when considered with [42] , represents the first comprehensive genetic characterization of a genetic background effect in any organism . Our work demonstrates how genetic background effects can arise due to complex epistatic relationships between mutations and cryptic variants at multiple modifier loci , as others have previously suggested [43] . Our findings also indicate that multiple epistatic configurations of cryptic variants may enable a given mutation to show a phenotypic effect . Although these results advance understanding of the causes of genetic background effects , determining the generality of these findings will require dissecting other genetic background effects that involve different mutations , species , and traits .
All phenotyping experiments were performed on agar plates containing yeast extract and peptone ( YP ) with 2% ethanol as the carbon source ( YPE ) . Prior to phenotyping , strains were grown to stationary phase in liquid YP with 2% dextrose ( YPD ) . Cultures were manually pinned onto YPE and allowed to grow for five days at 30°C , and were then imaged using a standard digital camera . Strains with opposite mating types were mixed together on a YPD plate and incubated for four hours at 30°C . A zygote from each cross was obtained by microdissection . To generate segregants , diploids were sporulated at room temperature using standard yeast sporulation procedures [66] . Once sporulation had completed , spore cultures were digested with β-glucuronidase and then plated onto YPE plates at a density of roughly 100 to 200 colonies per plate . Approximately 10 plates were screened per backcross . 148 ( BY backcross ) and 88 ( 3S backcross ) rough segregants were picked manually and streaked to obtain single cell isolates . The mating type of each of these strains was checked to confirm that they were indeed haploid . Segregants from the BY backcross could be either END3BY or END33S . In order to ensure sequenced strains possessed the END33S allele , each segregant was genotyped using a nearby restriction marker ( S1 Table ) . 63 of the 148 BY backcross progeny possessed the END33S allele and were used for genetic mapping . We note that other multicellularity phenotypes ( e . g . , flocculation ) segregated in the backcrosses , but were not strongly correlated with expression of the rough phenotype , implying they have different genetic architectures . The BY and 3S strains used in the ira2Δ2933 and sfl1Δ crosses possessed the Synthetic Genetic Array marker system [67] , which allowed for generation of large numbers of recombinant MATa progeny . Regarding the IRA2 wild type cross , we re-mated BY and 3S to produce a different diploid than the one used in [42] . For the ira2Δ2933 cross , the lesion was introduced into 3S using allele replacement techniques described below and then this 3S ira2Δ2933 strain was mated to a wild type BY strain . We designed the cross in this way because the ira2Δ2933 mutation originally occurred in the 3S allele of the gene . However , we note that we have never seen evidence for a genetic interaction between ira2Δ2933 and other genetic variants in IRA23S . As for the sfl1Δ cross , we constructed BY and 3S strains that lacked the entire coding region of SFL1 using genetic engineering techniques described below . A BY/3S sfl1Δ/sfl1Δ diploid was then used to generate a population of BYx3S sfl1Δ recombinants . For each of the three crosses described in this section , diploids were generated and sporulated as described for the backcrosses , but sporulations were plated at low density onto YNB plates containing canavanine to select for haploid progeny . These were then replica plated on YPE to phenotype colony morphology . For each cross , around 20 plates containing roughly 100 to 200 colonies were screened . Each rough END33S segregant from the backcrosses was grown to stationary phase as an individual , clonal culture . Cells from these stationary cultures were then mixed in equimolar fractions by backcross and DNA was extracted from the two pools using Qiagen G-tip columns . Whole genome sequencing libraries were prepared using the Illumina Nextera kit , with each of the backcross segregant pools barcoded with a unique sequence tag . The libraries were mixed together in equimolar fractions and sequenced on an Illumina MiSeq machine by the company Laragen , Inc . using 250 base pair ( bp ) x 250 bp reads . These sequencing reads were then mapped to the S . cerevisiae S288c reference and 322134S draft genomes ( http://www . yeastgenome . org ) . S288c is the progenitor of BY , and to ensure high quality read mapping , reads from the BY and 3S backcrosses were mapped to S288c and 3S , respectively . Alignments were performed using the Burrows-Wheeler Aligner ( BWA ) version 7 with options mem -t 20 [68] . Based on these alignments , we obtained 73- and 122-fold genomic coverage , as determined by the average per site coverages , from the BY and 3S backcross populations , respectively . A custom Python script was used to assess genome-wide allele frequencies at 36 , 756 high confidence SNPs that had previously been identified by mapping Illumina sequencing reads for 3S to the S288c genome [42] ( S2 Note; S2 Table ) . Loci influencing colony morphology were called as regions enriched at 95% frequency or higher when the data were averaged within running windows of 10 SNPs ( S2 Note ) . Intervals containing causal genes were identified in the R statistical programming environment as the smallest regions that had mean allele frequencies above a threshold of 95% ( S3 Note ) . Subsequent restriction typing experiments focused on individual segregants and the selected loci ( see S1 Table ) showed that the detected loci were in fact fixed , and that deviations from fixation occurred due to the presence of a small number of sequencing or read mapping errors . We note that Illumina data used for genetic mapping are available through the NCBI Sequence Read Archive under the study accession number SRP062432 , as well as the sample accession numbers SAMN03956543 ( BY backcross ) and SAMN03956544 ( 3S backcross ) . To generate allele replacement strains for ARP8 , LSC1 , MGA1 , SFL1 , SUF5 , and THI80 , a backcross segregant that expressed rough morphology due to the six-way genetic interaction was transformed using a modified form of adaptamer-mediated allele replacement [69] . Also , adaptamer-mediated allele replacement was used to introduce the ira2Δ2933 lesion into 3S . Transformations were conducted with two partially overlapping PCR products—a full-length amplicon of the gene of interest that was tailed at the 3’ end with the 5’ portion of the kanMX cassette and a copy of the kanMX cassette that was tailed on the 3’ end with part of the intergenic region downstream of the gene ( as shown in Figure S1 of [70] ) . Knock-ins were identified using selection on G418 and verified by Sanger sequencing . Deletions were constructed using the CORE cassette [71] . Homology tails matching the 60 bases immediately up- and downstream of each gene were attached to the CORE cassette through PCR and introduced into cells using the Lithium Acetate method [72] . Selection for G418 resistance was used to screen for integration of the CORE cassette; correct integration was then checked using PCR . SFL1 was deleted from BY and 3S , while FLO11 was deleted from a nearly isogenic line and a backcross segregant harboring the five- and six-way genetic interactions , respectively . All primers used for genetic engineering are provided in S1 Table . Markers within END3 , FLO8 , MGA1 , MSS11 , SFL1 , and TRR1 were genotyped using PCR and restriction digestion ( S1 Table ) . These markers were identified from among the 36 , 756 high confidence SNPs that differentiate BY and 3S . Strains were grown to stationary phase in liquid YPD media at 30°C and pinned on to YPE agar plates . After four days of growth at 30°C , total RNA was extracted with the Qiagen RNeasy kit . cDNA was then generated with Superscript reverse transcriptase from Life Technologies . ACT1 , a well-known housekeeping gene , was used as a control for our FLO11 RT-PCRs . Strains that were used in the RT-PCR experiments are described in the main text . The specific primers that we used were taken from [73] and are provided in S1 Table . | Some genetic polymorphisms have phenotypic effects that are masked under most conditions , but can be revealed by mutations or environmental change . The genetic and molecular mechanisms that suppress and uncover these cryptic genetic variants are important to understand . Here , we show that a single mutation in a yeast cross causes a major phenotypic change through its genetic interactions with two specific combinations of cryptic variants in six genes . This result suggests that in some cases cryptic variants themselves play roles in revealing their own phenotypic effects through their genetic interactions with each other and the mutations that reveal them . We also demonstrate that most of the genes harboring cryptic variation in our system are transcription factors , a finding that supports an important role for perturbation of gene regulatory networks in the uncovering of cryptic variation . As a final part of our study , we interrogate how a mutation exposes combinations of cryptic variants and obtain evidence that it does so by disrupting the silencing of one or more genes that must be expressed for the cryptic variants to exert their effects . To prove this point , we delete the transcriptional repressor that mediates this silencing and demonstrate that this deletion reveals a similar set of cryptic variants to the ones that were discovered in the initial mutant background . These findings advance our understanding of the genetic and molecular mechanisms that reveal cryptic variation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Transcriptional Derepression Uncovers Cryptic Higher-Order Genetic Interactions |
Human African trypanosomiasis ( HAT ) , also known as sleeping sickness , is a parasitic tropical disease . It progresses from the first , haemolymphatic stage to a neurological second stage due to invasion of parasites into the central nervous system ( CNS ) . As treatment depends on the stage of disease , there is a critical need for tools that efficiently discriminate the two stages of HAT . We hypothesized that markers of brain damage discovered by proteomic strategies and inflammation-related proteins could individually or in combination indicate the CNS invasion by the parasite . Cerebrospinal fluid ( CSF ) originated from parasitologically confirmed Trypanosoma brucei gambiense patients . Patients were staged on the basis of CSF white blood cell ( WBC ) count and presence of parasites in CSF . One hundred samples were analysed: 21 from stage 1 ( no trypanosomes in CSF and ≤5 WBC/µL ) and 79 from stage 2 ( trypanosomes in CSF and/or >5 WBC/µL ) patients . The concentration of H-FABP , GSTP-1 and S100β in CSF was measured by ELISA . The levels of thirteen inflammation-related proteins ( IL-1ra , IL-1β , IL-6 , IL-9 , IL-10 , G-CSF , VEGF , IFN-γ , TNF-α , CCL2 , CCL4 , CXCL8 and CXCL10 ) were determined by bead suspension arrays . CXCL10 most accurately distinguished stage 1 and stage 2 patients , with a sensitivity of 84% and specificity of 100% . Rule Induction Like ( RIL ) analysis defined a panel characterized by CXCL10 , CXCL8 and H-FABP that improved the detection of stage 2 patients to 97% sensitivity and 100% specificity . This study highlights the value of CXCL10 as a single biomarker for staging T . b . gambiense-infected HAT patients . Further combination of CXCL10 with H-FABP and CXCL8 results in a panel that efficiently rules in stage 2 HAT patients . As these molecules could potentially be markers of other CNS infections and disorders , these results should be validated in a larger multi-centric cohort including other inflammatory diseases such as cerebral malaria and active tuberculosis .
Human African trypanosomiasis ( HAT ) , also called sleeping sickness , is a parasitic disease that occurs in sub-Saharan Africa . More than sixty million people are at risk of being infected . The World Health Organization ( WHO ) has reported impressive progress since 1995 in the control of HAT , leading to a substantial reduction of new cases detected yearly to 10'800 in 2007 . The total number of cases is now estimated to be between 50'000 and 70'000 per year [1] . The parasite that causes HAT belongs to the Trypanosoma brucei family with two subspecies , Trypanosoma brucei gambiense and Trypanosoma brucei rhodesiense , responsible for the human disease . Trypanosomes are transmitted to humans by the bite of a tsetse fly and are initially confined to the blood , lymph nodes and peripheral tissues . This corresponds to the first stage ( early stage; or haemolymphatic stage ) of the disease . After an unknown period that varies from weeks to months , the parasites invade the central nervous system ( CNS ) . This is called the second stage ( late stage; or neurologic; or meningo-encephalitic stage ) of HAT . Clinical symptoms of HAT are not specific for the disease , and definite diagnosis is always based on parasitological examination of body fluids . The card agglutination test for trypanosomiasis ( CATT ) , an assay that is based on trypanosome-specific antibody detection , is widely used for mass screening . However , it suffers from limited sensitivity and restricted to the T . b . gambiense form of the disease [2] . A positive parasitological diagnosis must always be followed by stage determination , which is performed by examination of the cerebrospinal fluid ( CSF ) . This is a vital step in the diagnostic process , as the treatment differs depending on the stage of the disease . If HAT patients are not treated , they always die [3]–[5] . Early stage drugs are inefficient for late stage patients , and additionally , melarsoprol ( MelB or Arsobal ) , which has been the most widely used drug to treat late stage patient , has itself an overall mortality rate of 5% due to its toxicity [6] . As a consequence , melarsoprol has in many countries been replaced by eflornithine as the first line treatment for T . b . gambiense infections but the latter drug suffers from important logistic constraints . WHO defined late-stage HAT by the following criteria: presence of trypanosomes in CSF and/or an elevated WBC count above 5/µL of CSF [7] . However , presence of WBC in the CSF is not specific for the disease and parasite detection methods are not sensitive enough [8] . Furthermore , recent studies suggest the need to increase the cutoff between the first and second stages to 10 or 20 WBC/µL [2] , [8] , [9] . This has contributed to the concept of a potential intermediate stage of HAT with CSF WBC count >5 and ≤20 WBC/µL [10] . There is therefore a critical need for a reliable and efficient staging tool that would replace or complement trypanosome detection and WBC count . Parasite migration and invasion of the CNS causes a neuroinflammatory process , associated with activation of microglial cells and astrocytes [11] , [12] , and infiltration of the CNS with leukocytes ( predominantly mononuclear cells ) [13] . Cytokines and chemokines are known to be actively involved in this process . Thus , TNF-α , IL-6 , CXCL8 and IL-10 concentrations have been demonstrated to be elevated in the CSF of late-stage patients [11] , [14] and the IFN-γ level has been reported as associated with the severity of the late stage disease [15] . The levels of CCL2 , IL-1β and CXCL8 have also been correlated with presence of parasites in the CSF and neurological signs in HAT patients [16] . Additionally , levels of IL-1ra , G-CSF , VEGF , CCL4 and CXCL10 were found modulated in either the CSF or plasma of patients suffering from cerebral malaria [17]–[19] , and could potentially be also modulated in HAT patients . Proteomic analysis of human body fluids has become an important approach for biomarkers discovery [20] . In this context , we recently explored the concept of post-mortem CSF as a model of massive and global brain insult [21] , which allowed the identification of potential brain damage biomarkers by proteomics strategies . Indeed , heart-fatty acid binding protein ( H-FABP ) , identified from post-mortem CSF , has been validated as a marker of stroke [22] and Creutzfeldt-Jakob disease [23] , respectively . Similarly , GSTP-1 was also found over-expressed in post-mortem CSF [24] compared to ante-mortem , and was recently validated as an early diagnostic marker of stroke and traumatic brain injury ( Turck et al . Personal communication ) . Additionally , S100β protein has already been demonstrated to be a marker of blood-brain barrier ( BBB ) and neuronal damage [25] as well as a useful serum biomarker of CNS injury and a potential tool for predicting clinical outcome after brain damage [26] . In this context , we hypothesized that markers of brain damage discovered by proteomic strategies as well as inflammation-related proteins could individually or in combination indicate the CNS invasion by the trypanosome parasite . We measured the CSF concentrations of H-FABP , GSTP-1 , S100β and thirteen inflammation-related proteins ( IL-1ra , IL-1β , IL-6 , IL-9 , IL-10 , G-CSF , VEGF , IFN-γ , TNF-α , CCL2 , CCL4 , CXCL8 and CXCL10 ) and evaluated their potential for staging the disease .
Samples originated from a prospective observational study on shortening of post treatment follow-up in gambiense human African trypanosomiasis ( THARSAT ) , conducted between 2005 and 2008 at Dipumba hospital in Mbuji-Mayi ( Kasai Oriental province , Democratic Republic of the Congo ) . Details of the THARSAT study design and results are reported elsewhere ( D . Mumba Ngoyi , in preparation ) . The study protocol was approved by the Ministry of Health , Kinshasa , DRC and by the Ethical Committee of the University of Antwerp , Belgium . Briefly , 360 T . b . gambiense patients in total were enrolled into the THARSAT study . Inclusion criteria were 1° confirmed presence of trypanosomes in lymph nodes , blood or CSF; 2°≥12 years old and; 3° living within a perimeter of 100 km around Mbuji-Mayi . Exclusion criteria were 1° pregnancy; 2° no guarantee for follow-up; 3° moribund; 4° haemorrhagic CSF before treatment and; 5° presence of another serious illness ( active tuberculosis - treated or not , bacterial or cryptococcal meningitis ) . HIV and malaria were not considered as exclusion criteria . Each patient underwent a clinical examination . Staging of disease was based on CSF examination . WBC count was performed in disposable cell counting chambers ( Uriglass , Menarini ) and was performed in duplicate when the first count was <20 cells/µL . Trypanosomes were searched for in CSF by direct examination prior or during the cell counting procedure , followed by the modified single centrifugation method [27] . Second stage patients were defined as having >5 WBC/µL and/or trypanosomes in the CSF . First stage patients were defined as having 0–5 WBC/µL and no trypanosomes in the CSF . Patients having >5 and ≤20 WBC/µL and no trypanosomes in CSF were defined and treated as stage 2 patients , but highlighted as being in the potential intermediate stage . Patients or their responsible were informed about the study objectives and modalities and were asked to provide written consent . Treatment was provided according to the guidelines of the national control program for HAT ( PNLTHA ) . CSF samples were centrifuged immediately after collection . The supernatant remaining after the diagnostic procedure was aliquoted , stored and shipped frozen at −20°C or below . For the study reported here , a total of 100 CSF samples , taken before treatment , were tested . These samples originated from 21 stage 1 ( S1 ) and 79 stage 2 patients ( S2 ) . S1 patients were age and sex matched with 21 S2 patients . Remainder S2 patients were chosen in order to obtain homogenous median age values . Patients were classified into three categories of neurological signs; absent ( no neurological signs ) , moderate ( at least one major neurological sign but no generalised tremors ) or severe ( at least two major neurological signs including generalised tremors ) . Major neurological signs were defined as: daytime somnolence , sensory and gait disturbances , presence of primitive reflexes ( Babinski's sign , palmo-mental reflex , perioral reflex ) , modified tendon reflexes ( exaggeration or abolition ) , abnormal movements such as tremor ( fine , diffuse and generalised ) . Neurological signs were not reported for two patients . The concentration of S100β was measured using a commercially available sandwich ELISA assay kit ( Abnova , Taiwan ) following the manufacturer's instructions . Briefly , calibrators , Quality control ( QC ) and CSF samples diluted 1∶4 were incubated 2 hours on microtiter strips pre-coated with polyclonal anti-cow S100β antibodies . After 3 washes , horseradish peroxidase ( HRP ) labelled anti-human S100β antibodies were added , incubated for 90 minutes and washed again before addition of the substrate solution ( tetramethylbenzidine ) . Color development was stopped with sulphuric acid and absorbance was read on a Vmax Kinetic microplate reader , ( Molecular Devices Corporation , Sunnyvale , CA , U . S . A . ) at a wavelength of 450 nm . H-FABP concentration was also determined using a commercially available ELISA kit ( Hycult Biotechnology , Uden , Netherlands ) according to the manufacturer's instructions . CSF samples ( non-diluted ) and standards were incubated ( 1 hour ) together with peroxidase conjugated secondary antibodies in microtiter wells coated with antibodies recognizing human H-FABP . After 3 washes , tetramethylbenzidine was added and color development was stopped by adding citric acid . The concentration of GSTP-1 was determined using a homemade ELISA as described by Allard et al . [28] . Briefly , biotinylated anti-GSTP-1 antibodies ( 2 µg/mL ) ( Biosite , California , USA ) were coated onto a 96-well Reacti-Bind NeutrAvidin coated Black Plates ( Pierce , Rockford , IL ) for 1 hour at 37°C . After 3 washes , CSF samples ( diluted 1∶4 ) , quality controls and standards ( recombinant GSTP-1 at concentrations ranging from 0 to 100 ng/mL ) were incubated for 1 hour at 37°C , and followed by a washing step . Alkaline phosphatase conjugated antibodies against human GSTP-1 ( Biosite , California , USA ) at 2 µg/mL were added and incubated for 1 hour at 37°C . After 3 washes , Attophos AP fluorescent substrate ( Promega , Madison , WI ) was added and plates were read immediately on a SpectraMax GEMINI-XS ( Molecular Devices Corporation , Sunnyvale , CA , U . S . A . ) plate reader , using the kinetic mode . Vmax values were automatically calculated by the instruments based on relative fluorescence units ( RFU ) ( λexcitation = 444 nm and λemission = 555 nm ) . Concentrations of S100β , H-FABP and GSTP-1 in the CSF samples were back-calculated using a linear calibration curve based on measured standards values . The levels of thirteen cytokines and chemokines ( IL-1ra , IL-1β , IL-6 , IL-9 , IL-10 , G-CSF , VEGF , IFN-γ , TNF-α , CCL2 , CCL4 , CXCL8 and CXCL10 ) were determined using the Bioplex bead suspension arrays according to the manufacturer's instructions ( Bio-Rad , Hercules , CA ) . Briefly , thirteen sets of color-coded polystyrene beads were conjugated separately with one of the thirteen different antibodies against the molecule of interest . All the sets were then mixed together by the supplier and delivered ready-to-use . An equal amount of beads was added to each well of a 96-well filter plate . After a series of washes , standards and samples ( diluted 1∶4 ) were added and incubated for 30 minutes at room temperature . After washing , a mix of the corresponding thirteen biotinylated detection antibodies was added and incubated 30 minutes at room temperature . After washing , streptavidin-phycoerythrin ( streptavidin-PE ) was added for 10 minutes . After a last series of washes , beads were re-suspended in the provided assay buffer and each well was aspirated using the Bio-Plex system . Each bead was identified and the corresponding target simultaneously quantified based respectively on bead color and fluorescence . The concentration of each target was automatically calculated by the Bio-Plex Manager software using corresponding standard curve ( 5-PL regression ) obtained from recombinant protein standards . Descriptive statistics were performed using the SPSS ( version 16 . 0 , SPSS Inc . , Chicago , IL , USA ) and GraphPad Prism ( version 4 . 03 , GraphPad software Inc . , San Diego , CA , USA ) software . Because none of the markers presented a normal distribution in concentrations ( Kolmogorov-Smirnov test ) , differences between groups were tested with non-parametric Mann-Whitney U test ( comparison between two groups ) and Kruskal-Wallis test followed by Dunn's post-hoc test ( comparison between three groups ) . Statistical significance for these tests was set at 0 . 05 ( 2-tailed tests ) . The stage , the presence of the parasite in CSF and the severity of neurological signs were successively considered as the dependent variables . The different marker concentrations were considered as independent variables . Bivariate non-parametric correlations using the Spearman correlation coefficient were carried out with statistical significance set at 0 . 01 ( 2-tailed tests ) . To calculate the sensitivity and specificity of each individual predictor with respect to staging , the specific receiver operator characteristic ( ROC ) curve of each analyte was determined and the cutoff value was selected as the threshold predicting stage 2 patients with 100% of specificity ( Figure S1 ) . Aabel ( version 2 . 4 . 2 , Gigawiz Ltd . Co . , Tulsa , OK , USA ) was used for box plots , SPSS for scatter plots and R ( version 2 . 8 . 0 ) [29] was used for plotting ROC curves . Panel selection was mainly performed as described by Reynolds et al . [30] . Briefly , the optimized cutoff values were obtained by modified iterative permutation-response calculations ( rule-induction-like , RIL ) using only the molecules that presented a p value<0 . 0001 ( Mann-Whitney U test ) , an AUC above 75% and a significant Spearman correlation with WBC above 0 . 4 ( Table 2 ) . Each cutoff value was changed iteratively by quantile of 2% increment and sensitivity was determined after each iteration until a maximum sensitivity was achieved for 100% specificity . The permutation–response calculations were conducted using a PERL program ( ActivePerl version 5 . 10 . 0 . 1004 , ActiveState Software Inc . ) and data were coded in CSV format .
The main characteristics of the 100 patients evaluated in this study are presented in Table 1 . The analytes were classified into three groups , based on the results presented in Table 2 . Criteria for the classification were the significance ( Mann-Whitney U test ) , the AUC and the correlation with WBC . In the first group ( GR1 ) comprising IL-1ra , G-CSF , CCL4 , and VEGF , no significant difference in CSF concentrations between the two stages of HAT was observed . The second group ( GR2 ) encompassed IFN-γ , IL-9 , CCL2 and S100β , for which concentrations in the CSF were significantly different between stage 1 and stage 2 patients ( 0 . 001<p<0 . 01 , Mann-Whitney U test ) . The third group ( GR3 ) included GSTP-1 , H-FABP , TNF-α , IL-1β , IL-6 , IL-10 , CXCL8 and CXCL10 , for which the difference between stages was highly significant ( p<0 . 0001 , Mann-Whitney U test ) ( Figure S2 ) . To assess the sensitivity and specificity of these analytes for S2 HAT , ROC curves were built . GR1 and GR2 had a low to medium area under ROC curve ( AUC ) ranging from 54 to 70% and also displayed a low sensitivity in detecting S2 patients ( 4–13% for GR1 and 10–44% for GR2 , see Table 2 ) at a predefined specificity of 100% . GR3 showed higher AUC ( 79–95% ) , and sensitivities for identification of S2 patient up to 84% ( Table 2 ) . CXCL10 appeared then as the most accurate predictor for staging , as , with a cutoff set at 2080 pg/mL , this molecule identified 66 out of 79 late stage patients and ruled-out all the early-stage patients . As the white blood cell count was one of the two reference staging parameters , we investigated the correlation between the concentrations of the sixteen biomarkers and the number of WBC in CSF ( Table 2 ) . There was no significant correlation in the concentrations of the first and second group of analytes ( GR1 and GR2 ) with WBC , except for S100β , which had a significant but low Spearman rho coefficient ( 0 . 269 , p<0 . 01 ) . Otherwise , strong correlations were observed between WBC and the concentrations of GR3 biomarkers ( GSTP-1 , IL-1β , IL-6 , H-FABP , TNF-α , IL-10 , CXCL8 and CXCL10 ) , with Spearman rho ranging from 0 . 417 to 0 . 732 ( Table 2 and Figure 1 ) . The levels of GR3 molecules in 8 potential intermediate stage patients ( parasite not detected in CSF and having >5 and ≤20 WBC/µL ) demonstrated the intermediate behaviour of this category with some patients appearing as S1 and others as S2 patients ( Figure 1 ) . Based on the above results , only the GR3 molecules ( GSTP-1 , IL-1β , IL-6 , H-FABP , TNF-α , IL-10 , CXCL8 and CXCL10 ) were selected for further analyses . GR3 molecule concentrations were classified according to the absence/presence of trypanosomes in CSF . GSTP-1 , IL-1β , IL-6 , H-FABP , TNF-α , IL-10 , CXCL8 and CXCL10 concentrations were significantly increased in patients with parasites in CSF ( Figure 2 and Table S1 ) . The six biomarkers associated with inflammation had a lower p value ( <0 . 0001 , Mann-Whitney U test ) and higher AUC ( ranging from 78% to 89% ) than H-FABP and GSTP-1 ( 0 . 001<p<0 . 05 , Mann-Whitney U test , AUCs of 69% and 64% respectively ) . Additionally , when only S2 patients were analysed , CXCL10 , IL-10 and TNF-α levels still demonstrated a significant difference between patients with or without trypanosomes in CSF ( p<0 . 05 , Dunn's post-hoc test , Table S1 ) . The patients were classified with respect to the neurological signs reported ( absence , moderate or severe ) ( Figure 3 ) . All the GR3 molecules except GSTP-1 showed a significant increase in concentration associated with higher severity of neurological signs ( p<0 . 05 , Kruskal-Wallis test ) . Indeed , CXCL10 , CXCL8 , IL-6 , IL-10 , IL-1β , and TNF-α concentrations were significantly different between patients without neurological signs and severe neurological signs ( p<0 . 05 , Dunn's post-hoc test ) , as well as between patients with moderate and severe neurological signs ( p<0 . 05 , Dunn's post-hoc test ) . H-FABP level was significantly different between patients without neurological signs and severe neurological signs ( p<0 . 05 , Dunn's post-hoc test ) . Only the concentrations of CXCL10 , IL-10 and TNF-α could distinguish between absence and moderate neurological signs ( p<0 . 05 , Dunn's post-hoc test ) . In an effort to improve the global sensitivity of molecules in the prediction of second stage HAT , the GR3 molecules were combined using the rule induction like ( RIL ) approach . This resulted in the identification of a three-molecule panel characterized by CXCL10 , CXCL8 and H-FABP ( cutoff values were set at 2080 . 0 , 97 . 1 and 571 . 8 pg/mL , respectively ) . A positive test ( leading to identification of S2 patient ) was obtained as soon as one of the three molecules included in the panel was above its cutoff value ( Table 3 ) . The panel had a sensitivity of 97% ( 95% CI , 91–100% ) and , by definition , a specificity of 100% ( 95% CI , 84–100% ) . This means that the panel could identify 77 out of 79 stage 2 patients , and ruled-out all the 21 stage 1 patients . Out off the 77 ruled-in S2 patients , 5 were CXCL10 positive only ( >2080 . 0 pg/mL ) , 6 CXCL8 positive only ( >97 . 1 pg/mL ) and 3 H-FABP positive only ( >571 . 8 pg/mL ) . The rest of ruled-in S2 patients were identified with either 2 positive molecules ( n = 23 ) or 3 positive molecules ( n = 40 ) . When this panel was applied on the intermediate stage patients ( eight patients having >5 and ≤20 WBC/µL and no trypanosomes in CSF ) only one patient gave a negative test response and thus 7 out of 8 patients were classified as S2 .
In this study , including early and late stage HAT patients ( n = 100 ) , we evaluated sixteen molecules as potential staging markers of HAT , to replace or complement trypanosome detection and WBC count . Eight of these molecules , CXCL10 , CXCL8 , IL-6 , IL-10 , IL-1β , TNF-α , H-FABP and GSTP-1 , presented concentrations significantly elevated in the CSF of late-stage HAT patients . We demonstrated that the CSF concentration of CXCL10 is highly elevated in stage 2 patients when compared to stage 1 , highlighting this molecule as a potential new staging marker for sleeping sickness . A combinatorial approach has been applied in staging of HAT , in order to improve the sensitivity . This method has led to the identification of a panel consisting of CXCL10 , CXCL8 and H-FABP , that identified late-stage patients with a sensitivity of 97% at 100% specificity . H-FABP is a small protein belonging to the fatty acid-binding proteins ( FABPs ) and known to be expressed in the brain [31] . In myocardial infarction , HFABP is quickly released after the tissue damage [32] , [33] . It has been suggested that the release of H-FABP from damaged cells could be used for diagnosis of acute and chronic brain injuries [31] . GSTP-1 is a member of the Glutathione S-transferase superfamily , playing a role in oxidative stress . Its expression in brain has not been well studied , but GSTP-1 seems to be the main isoform in brain [34] and may function as a brain damage biomarker [24] . Our results showed a higher level of both H-FABP and GSTP-1 in CSF of late stage patients compared to early stage patients . These two molecules are known to be associated with early brain cell death [24] , [31] , which could be correlated with the observed increase of their concentration in late-stage HAT patients . From now , it is not know if these two molecules were also associated with the inflammatory process . Cytokines and chemokines play an important role in inflammatory processes and blood-brain barrier ( BBB ) dysfunction [35] , and could therefore be potentially used as markers for staging HAT [11] , [16] , [36] . In the present study , the measured levels of inflammation-related proteins in CSF showed significant differences according to the disease progression . Indeed , concentrations of IL-1β , IL-6 , IL-10 , TNF-α , CXCL8 and CXCL10 were increased in the CSF of patients in late stage HAT compared to those in early stage of the disease . In addition , the levels of IL-1β , IL-6 , CXCL8 and IL-10 were similar to those already reported for T . b . gambiense HAT [11] , [16] . IL-1β is a pro-inflammatory cytokine that induces leukocytes infiltration [37] and is rapidly expressed in response to brain damage [35] . The high level of IL-1β found in CSF of stage 2 patients confirmed its probable association with the inflammatory process . Furthermore , its level was clearly correlated to the presence of severe neurological signs , supporting a potential release in relation to neurodegeneration . IL-6 and IL-10 are both anti-inflammatory cytokines . Their increased level in the CSF according to the stage as well as the severity of the neurological signs confirmed their activation associated with disease progression . The concentration of the two molecules was significantly increased in patients with more than 20 WBC/µL , which may suggest a probable expression after an already activated inflammatory process . Indeed , it has been demonstrated in vervet monkey models of HAT that IL-10 is associated with down-regulation of pro-inflammatory cytokines ( IFN-γ and TNF-α ) in the late stage of T . b . rhodesiense disease [38] . The level of the pro-inflammatory chemokine CXCL8 was also significantly elevated in CSF of S2 patients and correlated well with both presence of trypanosomes in CSF and severity of neurological signs . CXCL8 is a strong neutrophil attractant [16] , which could thus not explain the good correlation of CXCL8 and the number of WBC ( mainly B-lymphocytes ) in CSF . However , its elevation in patients with a relatively low number of WBC ( between 5 and 20/µL ) suggests an early activation , which may play a role in BBB dysregulation [11] . The pro-inflammatory cytokine TNF-α has been reported as being involved in blood-brain barrier dysfunction [39] . These authors also demonstrated that trypanosomes may induce synthesis of TNF-α . In the present study , the increasing level of TNF-α was associated with disease progression as well as the presence of the parasite in CSF . These results suggested that parasites invasion into the CNS may lead to TNF-α production , which generated then CNS inflammation [14] . Additionally , an elevation according to the severity of the neurological symptoms was observed , which may support the neurotoxic effect of this cytokine in HAT [35] . CXCL10 , also known as IP-10 , is a pro-inflammatory chemokine with a central role in inflammatory responses [40] . The main effect of CXCL10 as a chemotactic molecule is activation of T cell migration to the site of inflammation , after binding to its receptor , CXCR3 [41] . The involvement of this chemokine in different CNS disorders has been demonstrated , such as viral meningitis [42] and multiple sclerosis [43] , where increased CXCL10 levels in the CSF correlated with tissue infiltration of T lymphocytes [44] . In our study , the concentration of CXCL10 increased with progression of the disease , and was highly correlated with the number of WBC in CSF . Many studies have pointed out astrocytes as the primary source of CXCL10 at the level of the CNS and showed that this molecule is responsible , as chemoattractant , for the influx of activated T lymphocytes in brain [43] , [45]–[47] . Indeed , there is a predominance of plasma cell infiltration in the brain of trypanosomiasis infected individuals . In addition , it has very recently been shown in a mouse model of HAT that CXCL10 may play an important role in T-cell recruitment into the brain parenchyma and is probably associated with brain invasion by trypanosomes [48] . Furthermore , the early activation of cytokine production ( TNF-α , IL-6 , and IFN-γ ) by astrocytes and microglia in mice models infected with T . brucei before observation of an inflammatory response [49] has confirmed an important role of astrocyte activation in CNS inflammatory response . In consequence , early astrocyte activation , which induces CXCL10 production , is probably linked with BBB dysfunction and may occur before the inflammatory process . These hypotheses were supported by the increase CXCL10 concentration observed in patients having >5 and ≤20 WBC/µL but without trypanosomes detected in the CSF . The CXCL10 level was also demonstrated to be elevated in patients with cerebral malaria , and pointed out as potentially inducing apoptosis of endothelial cells leading to BBB breakdown [17] Recent work has suggested that neuronal apoptosis associated with calcium dysregulation may be induced by CXCL10 [50] . Even if mechanisms of CXCL10 mediated neurotoxicity remain unclear , we showed that the concentration of CXCL10 was correlated to the severity of neurological signs , supporting a possible involvement of this protein in neuronal injury pathways . Thus , CXCL10 expression in late stage HAT patients may be associated with both cell death and inflammatory process . Finally , active tuberculosis and pregnancy , two exclusion criteria in this study , have also been reported as modulating the level of CXCL10 [51] , [52] . Although they have only been evaluated on serum and whole blood samples so far , it is not excluded that these criteria could potentially induce CXCL10 modulation in CSF . Nevertheless , our data demonstrated that CXCL10 is an efficient tool for staging patients , and suggested a potential role of CXCL10 as an early marker of parasite invasion into the CNS . As the investigated proteins may be involved in different biological mechanisms , we evaluated in this study a strategy to combine results of each molecule , in order to find a panel able to discriminate more accurately early and late stage patients . This highlighted a panel of three molecules , including CXCL10 ( the most promising single molecule ) , CXCL8 ( another chemokine ) and H-FABP ( a marker of brain damage ) . With a specificity of 100% , this panel increased the sensitivity for staging of HAT patients up to 97% ( compared to the 84% obtained with CXCL10 taken individually ) . Although the number of “intermediate” patients was small , the panel appeared to classify them rather as S2 patients ( 7/8 patients ) . This supports the current recommendation by WHO to consider such patients as S2 patients and treat them with drugs used for late stage disease . However , there is a need for more studies on T . b . gambiense and T . b . rhodesiense patients , before and after treatment , as well as on other parasitic diseases such as cerebral malaria , to verify these results and assess the feasibility of using the three-molecule panel as a complement to WBC count . There are obviously some drawbacks concerning this approach . Firstly the obtained panel is not 100% sensitive and thus some stage 2 patients will not be detected . The influence of other possible co-infections should also be evaluated in order to determine if they significantly modulate the evaluated molecules . Indeed , the three molecules included in the panel could potentially all be markers of other CNS disorders . It is also evident that the methods described in this study could not be implemented in such a way directly in the field and should be first transformed into a more simplified technique as for example a lateral flow immunoassay . Another limitation is the continued requirement of the invasive lumbar puncture since the molecules highlighted in this study have been evaluated on CSF samples . In conclusion , the present study demonstrated the utility of inflammation-related proteins and brain damage markers as indicators of the second stage of HAT but potentially in other CNS disorders as well . We highlighted the value of CXCL10 as an efficient staging biomarker for T . b . gambiense infected HAT patients . Additionally , a combination of CXCL10 with CXCL8 and H-FABP resulted in a highly sensitive tool for identification of late stage HAT patients . | The actual serological and parasitological tests used for the diagnosis of human African trypanosomiasis ( HAT ) , also known as sleeping sickness , are not sensitive and specific enough . The card agglutination test for trypanosomiasis ( CATT ) assay , widely used for the diagnosis , is restricted to the gambiense form of the disease , and parasitological detection in the blood and cerebrospinal fluid ( CSF ) is often very difficult . Another very important problem is the difficulty of staging the disease , a crucial step in the decision of the treatment to be given . While eflornithine is difficult to administer , melarsoprol is highly toxic with incidences of reactive encephalopathy as high as 20% . Staging , which could be diagnosed as early ( stage 1 ) or late ( stage 2 ) , relies on the examination of CSF for the presence of parasite and/or white blood cell ( WBC ) counting . However , the parasite is rarely found in CSF and WBC count is not standardised ( cutoff set between 5 and 20 WBC per µL ) . In the present study , we hypothesized that an early detection of stage 2 patients with one or several proteins in association with clinical evaluation and WBC count would improve staging accuracy and allow more appropriate therapeutic interventions . | [
"Abstract",
"Introduction",
"Material",
"and",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"biochemistry/bioinformatics"
] | 2009 | A Combined CXCL10, CXCL8 and H-FABP Panel for the Staging of Human African Trypanosomiasis Patients |
Leprosy is an insidious infectious disease caused by the obligate intracellular bacteria Mycobacterium leprae , and host genetic factors can modulate the immune response and generate distinct categories of leprosy susceptibility that are also influenced by genetic ancestry . We investigated the possible effects of CYP19A1 [rs11575899] , NFKβ1 [rs28362491] , IL1α [rs3783553] , CASP8 [rs3834129] , UGT1A1 [rs8175347] , PAR1 [rs11267092] , CYP2E1 [INDEL 96pb] and IL4 [rs79071878] genes in a group of 141 leprosy patients and 180 healthy individuals . The INDELs were typed by PCR Multiplex in ABI PRISM 3130 and analyzed with GeneMapper ID v3 . 2 . The NFKβ1 , CASP8 , PAR1 and IL4 INDELs were associated with leprosy susceptibility , while NFKβ1 , CASP8 , PAR1 and CYP19A1 were associated with the MB ( Multibacilary ) clinical form of leprosy . NFKβ1 [rs28362491] , CASP8 [rs3834129] , PAR1 [rs11267092] and IL4 [rs79071878] genes are potential markers for susceptibility to leprosy development , while the INDELs in NFKβ1 , CASP8 , PAR1 and CYP19A1 ( rs11575899 ) are potential markers for the severe clinical form MB . Moreover , all of these markers are influenced by genetic ancestry , and European contribution increases the risk to leprosy development , in other hand an increase in African contribution generates protection against leprosy .
Leprosy is an insidious infectious disease caused by the obligate intracellular bacteria Mycobacterium leprae that affects the skin and peripheral nerves , causing a chronic granulomatous infection [1] . Leprosy patients may be classified in two major groups , based on clinical manifestations using a simple system introduced by the WHO ( World Health Organization ) in 1982 . Paucibacillary ( PB ) is the primary characteristic of Tuberculoid ( TT ) leprosy and is characterized by a few lesions and scarce bacilli , and Multibacillary ( MB ) is the primary characteristic of anergic Lepromatous ( LL ) leprosy . From an epidemiological perspective , the situation in Brazil is critical because , along with India and Indonesia , it has the highest rate of new cases detected worldwide [2 , 3 , 4] . In addition to the system introduced by WHO in 1982 , the use of histological and immunological criteria as described by Ridley-Jopling further improves definition of Borderline cases . According to this classification , TT ( tuberculoid-tuberculoid ) patients , who have the PB type , exhibit a strong cellular immune response ( CIR ) mediated by Th1 , and a negative skin smear test . In contrast , LL ( lepromatous-lepromatous ) patients have a weak or absent CIR and a highly positive skin smear associated to an humoral immune response . In the middle of this spectrum are a large number of borderline patients , which together with LL comprise the MB pole , with symptoms varying from weak to strong CIR and negative to positive skin smears [5 , 6] . The regulation of the host immune response and manifestation of disease clinical between types PB ( better ) and MB ( severe ) involves cytokine and others mediators produced by various subtypes of T cells . In PB , an inflammatory immune response is mediated by Th1 cells that express pro-inflammatory interleukins that stimulate macrophages and phagocytosis mechanisms to inhibit bacillary growth and kill mycobacteria [2 , 7–9] . On the other hand , MB patients have an intense Th2 immune response with production of anti-inflammatory cytokines in addition to the specific anti-PGL-1 ( phenolic glycolipid 1 ) antibody . This mechanism does not block bacillary growth and contributes to the host’s inability to resist the development of severe disease [2 , 8 , 9–11] . Recent studies have investigated genetic markers , usually innate immune response genes , as possible susceptibility factors for leprosy because the SNPs in these genes can modulated the host immune response and consequently lower host resistance to bacillus growth [6 , 12 , 13] . However , few studies have investigated INDEL polymorphisms ( insertion-deletion ) in immune response genes in leprosy . Moreover , such polymorphisms present interesting features as genetic markers because i ) INDELs are spread throughout the human genome , ii ) INDELs derive from a single event ( they do not present homoplasy ) , iii ) small INDELs can be analyzed using short amplicons , which improves amplification of degraded DNA and facilitates multiplexing reaction , iv ) INDELs can create abrupt changes in the normal function of the gene and v ) INDELs can be easily genotyped using a simple dye-labeling electrophoretic approach [14] . The current study select eight INDEL in seven genes ( CYP19A1 , NFKβ1 , IL1α , CASP8 , UGT1A1 , PAR1 , CYP2E1 , and IL4 ) , which have relation with the immune response modulation in leprosy patients , beside literature that demonstrate these molecular markers like functional polymorphisms that alter transcriptional activity of the gene , and consequently the immunological phenotype against the bacilli . Additionally these INDELs can be able to contribution to construction a possible panel of susceptibility markers . However , from the genetic point of view , Brazil is recognized as having one of the most heterogeneous populations in the world , with important genetic information being contributed by three main continental groups , Europeans , Africans and Amerindian , resulting in a genetically very diverse modern Brazilian population [15] . Therefore , analysis of genetic markers in complex diseases may result in spurious results due to population substructure [16] , and it is important to perform the genomic ancestry control , especially in populations with a high degree of interethnic admixture [14] . The objective of this study was to investigate eight INDEL polymorphisms in seven genes involved in modulation of the host immune response , including CYP19A1 [rs11575899] , NFKβ1 [rs28362491] , IL1α [rs3783553] , CASP8 [rs3834129] , UGT1A1 [rs8175347] , PAR1 [rs11267092] , and CYP2E1 [INDEL 96pb] , besides one VNTR ( variable number tandem repeat ) of 70 bp on intron 3 of IL4 [rs79071878] in a group consisting of 141 leprosy patients and 180 healthy individuals , to identify possible susceptibility markers of leprosy and evaluate the influence of genetic ancestry on disease risk .
The project was approved by the Pará Federal University ethics committee ( N° 197/07 ) . We investigated 141 leprosy patients who attended the Dr Marcello Candia Reference Unit in Sanitary Dermatology of the State of Pará ( UREMC ) , in Marituba , Pará , Brazil between January 2008 and December 2009 . All patients were informed about the study before they signed informed consent forms . Since 2002 , UREMC registered between 308 and 472 leprosy patients ( mean: 408 cases per year ) . Of the 765 leprosy cases registered in 2008 and 2009 alone , 141 ( 18 . 43% ) were randomly selected for this study . These patients were divided according to Ridley-Jopling classification [5] into Paucibacillary ( TT: PB 31 ) and Multibacillary ( BT , BB , BL and LL: MB 110 ) groups . A total of 180 healthy individuals who were unrelated , without leprosy or other chronic diseases and from the same geographic area as each other were chosen for the control group . Leprosy patient’s descriptions were made previously [6] . These subjects were asked to participate in the study after being informed about the study objectives and signing informed consent forms . DNA extraction was performed as previously described by phenol-chloroform method [6 , 17] . The DNA concentration was determined by spectrophotometry ( Themo Scientific NanoDrop 1000 , NanoDrop Technologies , Wilmington , US ) . Individual interethnic admixture was estimated using a panel of 48 ancestry informative markers ( AIMs ) as previously described [6 , 14] . The allelic frequencies between healthy individuals and leprosy patients and between PB and MB patients were estimated by gene counting . Deviation from the Hardy-Weinberg equilibrium was assessed using chi-squared tests , using the Arlequin v3 . 5 software [18] , and p-value of HWE was corrected by Bonferroni methods . Differences between leprosy patients and healthy individuals and between PB and MB patients with respect to age , gender and genetic ancestry were estimated using Student’s t-Test , Fisher’s exact test and Mann-Whitney tests , respectively . The association of markers between groups was analyzed by logistic regression tests , all the test were corrected by FDR ( False Discovery Rate ) method , and all tests were performed using the statistical package under R calculation . A two-tailed p-value < 0 . 05 was considered statistically significant . The individual contributions of European , African and Amerindian genetic ancestry were estimated using the STRUCTURE 2 . 3 . 3 program assuming three parental populations ( European , African and Amerindian ) , a burn-in period of 200 , 000 , and 200 , 000 Markov Chain Monte Carlo repetitions after burn-in [16] . The differences in allelic frequencies between leprosy cases and the healthy individuals for markers analyzed following an adjustment for population stratification was performed using the STRAT software program with 10 , 000 simulations [16] .
The data of clinical and demographic distribution of leprosy patients and healthy individuals is shown in Table 1 . The mean age was higher in healthy individuals ( 55 . 7±12 versus 43 . 3±21 , p<0 . 001 ) , and male patients were more frequent among leprosy patients ( 97 [68 . 8%] versus 65 [36 . 1%] , p<0 . 001 ) . Analysis of ethnicity showed that the mean frequency of Africans was higher among leprosy patients ( 0 . 284 versus 0 . 236 , p<0 . 001 ) and Europeans were more frequent in healthy individuals ( 0 . 461 versus 0 . 427 , p = 0 . 004 ) . The frequencies of INDELs for the eight ( 8 ) genes analyzed in leprosy patients and healthy individuals are show in Table 2 . For the polymorphism in IL4 ( VNTR of 70 bp ) , only two alleles were identified in the sample . One allele had two repeats of 70 bp ( allele A1 ) and the other had three repeats of 70 bp ( allele A2 ) , suggesting theses alleles are biallelic markers . All the polymorphisms analyzed were according to the Hardy Weinberg equilibrium , therefore the association analysis were performed with regression logistic test and differences in allelic frequencies were corrected by frequencies of ancestry markers informative . When the INDELs were analyzed by logistic regression , the genes NFKβ1 and PAR1 showed statistically significant differences associated with the presence of the DEL allele ( p = 0 . 016 and p = 0 . 022 , respectively ) and both were associated like protection factors to not developing the disease ( OR[IC95%] = 0 . 50[0 . 27–0 . 88] and OR[IC95%] = 0 . 35[0 . 14–0 . 86] , respectively ) , for these genes was found a dominance effect DEL allele , that increase your protection capacity in general population . The CASP8 showed significant differences associated with the presence of the DEL/DEL homozygous genotype and was associated with a risk factor for leprosy development ( p = 0 . 017; OR[IC95%] = 2 . 33[1 . 16–4 . 69] ) ( Table 3 ) . The analysis of allele frequency differences was then corrected for the influence of genetic ancestry on population structure , and the results showed that the DEL allele of PAR1 gene and the allele A1 of IL4 is more frequent in healthy individuals ( p = 0 . 018 and p = 0 . 019 , respectively ) ( Table 3 ) , these results shown the importance of statistical correction in admixture population , in order to exhibit differences covert by structure population . Table 4 summarizes the clinical and demographic characteristics of leprosy patients grouped according to clinical manifestation in PB ( Paucibacillary ) and MB ( Multibacillary ) groups , and the only significant difference was observed for age ( p = 0 . 003 ) , with a higher mean age in MB patients ( 45 . 7±22 versus 34 . 9±15 ) . When the INDELs were analyzed by logistic regression , NFKβ1 showed significant differences like risk factor associated with the presence of the allele DEL in MB patients ( p = 0 . 024; OR[IC95%] = 2 . 64[1 . 13–6 . 19] ) , of contradictory way the dominance effect of DEL allele seem protect against the development of leprosy , but when the disease is established your effect seem inefficient to combat to bacilli . PAR1 showed significant differences associated with the presence of homozygous DEL/DEL genotype in PB patients ( p = 0 . 031; OR[IC95%] = 0 . 41[0 . 17–0 . 96] ) ( Table 5 ) . The analysis of allele frequency differences were corrected for population structure and showed that the DEL allele of CASP8 is more frequent in PB patients ( p = 0 . 003 ) , while the DEL allele of CYP19A1 is more frequent in MB patients ( p = 0 . 007 ) ( Table 5 ) . Fig 1 shows the OR ( odds ratio ) values obtained from leprosy patients and healthy individuals within groups having distinct level of ancestry composition . The figure shows that greater frequency of European ethnic between the groups ( leprosy patients and healthy individuals ) , higher is the risk for developing leprosy , while the smaller the frequency of the African ethnic , lower is the risk for developing leprosy . No statistically significant values were obtained for the analysis of the Amerindian group . These results are better understood on frequencies distribution , according with range of ancestry contribution ( S1 Table ) . For African ancestry 99 . 4% of health individual is closed between 0% and 50% of African contribution ( range that have p<0 . 05 on Fig 1 ) , moreover the contribution range of 10% to 30% is closed 81 . 7% of health individual , in this range the Fig 1 have more decline of OR value , that showed the higher protection effect of African ancestry . To European ancestry , 61 . 7% of leprosy patients is closed between 40% to 80% of European contribution , while 87 . 2% of health individuals is closed between 0% to 50% of European contribution . Additionally , for the contribution range between 60% to 80% we observed 17% of all patients , while no healthy individual was observed this range , these data show that leprosy patients have higher European contribution compared with healthy individuals . Take together the Fig 1 and S1 Table shown that to leprosy patients of an admixture population , like Brazil , African ethnic generates protection against the development of disease , and the opposite is also truth for European ethnic .
NF-κB belongs to family of protein transcription factors that modulate many inflammatory processes . In the resting state , IκBα ( inhibitor of NF-kβ activity ) sequesters NF-κB in the cytoplasm and prevents its activity , but in response to specific stimuli , IkBα is ubiquitinated and degraded allowing NF-kB to migrate to the nucleus and stimulate the transcription of proinflammatory genes [19 , 20] . The allele DEL ( rs28362491 ) has been shown to be associated with a decrease of transcriptional activity of variety genes of immune response [21] and with auto immune disease such as Systemic Sclerosis [22] and lupus erythematosus [23] . The role of NF-kβ in leprosy is not clear , and studies linked to expression of NF-kβ have suggested that lower expression is common in leprosy patients [24 , 25] . Our results suggest that the DEL carries genotype induces protection against leprosy ( Table 3 ) , although a comparison of PB and MB patients also suggests that DEL behaves like a risk factor for the development of the severe clinical form of MB ( Table 5 ) . Because the transcription of NF-kβ is mediated by specific stimuli , such as the presence of M . leprae [24] , it is conceivable that the presence of DEL confers risk to MB leprosy . PAR1 is a receptor of the PAR family of proteins that belong to a unique group of G protein—coupled receptors . In particular , PAR1 protein is present in a variety of cells like platelets , endothelia , epithelial , neurons , fibroblasts , smooth muscle , leukocytes and tumor lines [26] . This receptor has been shown to be involved in many natural physiological processes , that involve inflammation like the systems cardiovascular , respiratory and central nervous and in embryogenesis , cancer and inflammation [27] . PAR1 suppresses T helper type 1 ( Th1 ) and T helper type 17 ( Th17 ) cells and the secretion of IL-12 and IL-23 , thereby resulting in the inhibition of pro-inflammatory responses [28] . The allele of insertion ( INS ) of INDEL studied ( rs11267092 ) has been shown to increase gene transcription [29] and therefore , it is a risk allele for leprosy . Our results suggest that the presence of DEL induces protection against leprosy ( Table 3 ) , and the DEL/DEL genotype confers protection against the development of clinical forms of MB ( Table 5 ) , thus this genotype of PAR1 gene can suppresses cellular infiltration and increase both Th1 and Th17 responses to infection . Moreover , analyses of macrophages revealed that secretion of IL-12 and IL-13 , two cytokine that play role key on cellular immunity Th1 and Th17 , can be suppressed by PAR1 activation . Furthermore , PAR1 can suppress interferon regulatory factor 5 ( IRF5 ) , that play role key like transcription factor for IL-12 and IL-23 , which modulates the sub sets of cellular immunity . Thereby the suppression of IRF5 and IL-12/23 secretion by PAR1 gene , can provides a novel mechanism by which the host suppresses the Th1 and Th17response to infection , and dysregulation of this process can likely an important factor in the susceptibility of some individuals to leprosy [28] . Macrophages with a high load of M . leprae have been shown to undergo apoptosis , and this mechanism is under the control of cytokines [30] . In leprosy patients , the immune system is overburdened with bacilli , and most likely the continuous activation of T cells by circulating M . leprae antigens leads to apoptosis and to a reduction of peripheral lymphocytes and other immune effector cells in these patients with the regulation of apoptosis involved in the stimulation and activation of caspase-8 [31] . The allele DEL ( rs3834129 ) cause a decrease in CASP8 transcription and a reduction in apoptosis [32] , thereby improving the bacillary load . Our results suggested that the DEL/DEL genotype ( Table 3 ) and the high frequency of DEL allele ( Table 5 ) can raise the bacillary load and thus confers a risk to leprosy development . Interleukin-4 ( IL-4 ) is a key cytokine secreted by Th2 lymphocytes , eosinophils and mast cells that induces the activation and differentiation of B cells and the development of the Th2 subset of lymphocytes , which is ineffective in combating leprosy [33] . Our analysis of the VNTR on intron 3 of the IL4 gene ( rs79071878 ) revealed two common alleles with two ( A1 ) and three ( A2 ) tandem repeats . Of these , A2 allele is known to be a high producer of IL-4 [34] . Our results indicate that allele A2 is more frequent in leprosy patients compared to healthy individuals , consistent with the fact that higher levels of IL4 would be ineffective in controlling the growth of bacilli ( Table 3 ) . The conversion of androgens to estrogens , catalyzed by aromatase encoded by the CYP19A1 gene , is the primary pathway of estrogen production in humans [35] . The levels of these hormones are important in leprosy patients and it has been demonstrated that androgen levels are significantly lower in leprosy patients compared to healthy control subjects [36] . Moreover , there is an inverse correlation between plasma androgen levels and secretion of inflammatory cytokines , suggesting that high plasma androgen levels can be less effective in inhibiting bacillus growth [37] . The DEL allele ( rs11575899 ) has previously been reported to have a negative effect on aromatase activity [38] , and our results show that the DEL allele is more frequent in MB patients ( Table 5 ) . We hypothesize that the DEL allele can decrease aromatase activity and increase androgen levels , resulting in an overall reduction in effective combat of bacillary growth and development of the severe clinical form MB . It is unclear whether leprosy originated in Asia or Africa . However , leprosy is believed to have been introduced into Europe from India , and the incidence was high in Europe during the Middle Ages until approximately 1870 when the number of cases dramatically reduced because of socioeconomic development [39 , 40 , 41] . It is believed that leprosy was introduced in Brazil primarily by the Spanish and Portuguese [41] . Estimates indicate that before the arrival of colonizers , approximately 2 . 5 million natives lived in Brazil , and during the European immigration in the first three centuries , approximately 500 , 000 individuals came from Portugal and approximately 3 . 5 million Africans were brought into Brazil through slave trade [14] . Therefore , there is evidence of a so-called directed admixture process involving predominantly European , Native American and African people [42–45] . Our data indicates that the contribution of different ethnic groups to the composition of the current Brazilian population can generate different rates of risk for leprosy development according to the level of inter-ethnic composition of the individuals involved . Our analysis suggests that an increase in European contribution increases the risk of leprosy development , while an increase in African contribution decreases the risk for leprosy development and the Amerindian contribution does not result in any statistically significant differences ( Fig 1 ) . The introduction of leprosy in Brazil primarily can it be accredited to the slave trade , but no only for this reason . Slaves were firstly there from Africa , and in succeeding years the number these slaves were increased , but was not common between they the clinical manifestation of leprosy , because these slaves were from region of the Africa where leprosy was comparatively rare . Moreover isn't doubt that the Portuguese and , to a less degree , Dutch , French and Spaniards were responsible by introduction of leprosy in Brazil , on period of country colonization . Additionally , data showed that as early , as 1419 , the disease was common in Portuguese and epidemiologically in this time the leprosy was very prevalent in Europe , and particularly in Portugal [41] . Therefore , our data of risk of leprosy according the different ethnic groups compositions is consistent with the higher numbers of settlers Portuguese that came to Brazilian that probably increases the frequencies of alleles of susceptibility on Brazilian population [14 , 42–45] . In other hand , the African contribution may have increase the frequencies of allele that confer protection against to leprosy . Comparative analyses of the four M . leprae genomes ( India , Thailand , Brazil and US ) have revealed little clonal differences . Thus , the patterns of global human migration routes , during the past 100 , 000 years , corroborate and suggest that leprosy probably originated in Africa [46] . African-descendants in admixture populations can be less susceptible to the leprosy bacilli , probably because of genetic polymorphisms accumulated during these times , in gene that can modulate the immune response on infection combat . Furthermore African humans are the more genetically diverse population in the world consequently , by selection bias , genetic polymorphisms accumulated that confer protection against disease , can be present in this population and your descendants . Of point view epidemiological , the situation of African and Americas region is critical , and is associated the socioeconomic challenges related to the disease , but genetics components also are important to disease knowledge [4] . Thus understanding of like genetic ancestry , in admixture population , can to influence genetic susceptibility is essential to avoid spurious results . In conclusion , our study shows that the NFKβ1 [rs28362491] , CASP8 [rs3834129] , PAR1 [rs11267092] and IL4 [rs79071878] genes are possible markers for the susceptibility to development of leprosy and the severe clinical form MB . Moreover , after correcting for population structure within an admixture population , the results show that different levels of ethnic group composition can generate different OR rates for leprosy susceptibility . | Leprosy is an infectious disease caused by Mycobacterium leprae , which can carry to skin lesions and affect peripheral nerves , which cause physical and motor injuries on the patients . Moreover , leprosy , may be classified in two major groups , based on clinical manifestations in Paucibacillary ( PB ) or Multibacillary ( MB ) , and these phenotype may be influenced by host immune response; that can be controlled by genetics factors that can be useful like future panel of biomarkers to leprosy , and it’s related with the different genetic background of population studied . Therefore , we conducted a study to evaluate seven INDEL polymorphisms in seven genes involved in modulation of the host immune response , and consequently can modulated o phenotype showed through the disease , to identify possible susceptibility markers of leprosy . However this analysis can be spurious on presence of population structure , common in admixture population like the Brazilian , thus we evaluate like the influence of genetic ancestry can modulated the disease risk . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Influence of Genetic Ancestry on INDEL Markers of NFKβ1, CASP8, PAR1, IL4 and CYP19A1 Genes in Leprosy Patients |
We address the problem of homology identification in complex multidomain families with varied domain architectures . The challenge is to distinguish sequence pairs that share common ancestry from pairs that share an inserted domain but are otherwise unrelated . This distinction is essential for accuracy in gene annotation , function prediction , and comparative genomics . There are two major obstacles to multidomain homology identification: lack of a formal definition and lack of curated benchmarks for evaluating the performance of new methods . We offer preliminary solutions to both problems: 1 ) an extension of the traditional model of homology to include domain insertions; and 2 ) a manually curated benchmark of well-studied families in mouse and human . We further present Neighborhood Correlation , a novel method that exploits the local structure of the sequence similarity network to identify homologs with great accuracy based on the observation that gene duplication and domain shuffling leave distinct patterns in the sequence similarity network . In a rigorous , empirical comparison using our curated data , Neighborhood Correlation outperforms sequence similarity , alignment length , and domain architecture comparison . Neighborhood Correlation is well suited for automated , genome-scale analyses . It is easy to compute , does not require explicit knowledge of domain architecture , and classifies both single and multidomain homologs with high accuracy . Homolog predictions obtained with our method , as well as our manually curated benchmark and a web-based visualization tool for exploratory analysis of the network neighborhood structure , are available at http://www . neighborhoodcorrelation . org . Our work represents a departure from the prevailing view that the concept of homology cannot be applied to genes that have undergone domain shuffling . In contrast to current approaches that either focus on the homology of individual domains or consider only families with identical domain architectures , we show that homology can be rationally defined for multidomain families with diverse architectures by considering the genomic context of the genes that encode them . Our study demonstrates the utility of mining network structure for evolutionary information , suggesting this is a fertile approach for investigating evolutionary processes in the post-genomic era .
Accurate identification of homologs , sequences that share common ancestry , is essential for accuracy in function prediction and comparative genomics . Homology identification is integral to the annotation of novel genes [1] and prediction of gene function by various methods , including phylogenetic clustering [2] , gene fusion analysis [3] , [4] , phylogenomic inference [5] , and genomic context [6] , [7] . Homologous genes are used as markers to identify homologous chromosomal regions for comparative mapping [8] , [9] , analysis of whole genome duplication [10] , [11] , phylogenetic footprinting [12] , and operon prediction [13]–[15] . Pairwise homology detection is also an integral component of clustering approaches to protein family classification ( [1] , [16] , and work cited therein ) . All of these applications exploit one or both of the following properties of homologous sequences: genes that share common ancestry tend ( 1 ) to have similar structure and function , and ( 2 ) be located in homologous chromosomal regions , making them suitable markers for comparative genomics . Because of their prevalence and importance , it is desirable to incorporate multidomain sequences in such analyses: Multidomain proteins represent 40% of the metazoan proteome , with functional roles in signal transduction , cellular adhesion , tissue repair , and immune response [17] . However , multidomain sequences , especially those with promiscuous domains that occur in many contexts , are frequently excluded from genomic analyses due to lack of a theoretical framework and practical methods for detecting multidomain homologs . In this paper , we extend the traditional definition of homology [18] to multidomain sequences that share a common ancestral gene , providing a formalism suitable for modeling multidomain family evolution , design and validation of multidomain homology identification methods , and incorporation of multidomain sequences in genomic analyses . The original definition of molecular homology [18] does not capture multidomain evolution . Homology traditionally refers to evolution from a common ancestor by vertical descent ( e . g . , gene duplication and speciation ) , but multidomain proteins evolve via both vertical descent and domain insertion . For example , Figure 1 depicts two genes , a and b , which share not only a homologous domain but also a common ancestral gene . In contrast , b and c are a domain-only match , a pair of sequences that share similarity due to insertion of the same domain into both sequences but are otherwise unrelated . Beta platelet-derived growth factor receptor ( PDGFRB ) and cGMP-dependent protein kinase 1 , beta ( PRKG1B ) , in Figure 2A , are enzymes involved in protein amino acid phosphorylation and provide a concrete example of this situation . Phylogenomic and structural evidence [19]–[22] , as well as the promiscuity of the Ig and cNMP-binding domains , supports the common ancestry of this pair ( see Methods ) . They have a statistically significant alignment with an E-value of 2 . 4e−8 that covers 13% of the average of their lengths . While they share a common domain ( Pkinase ) , the Ig domains are unique to PDGFRB and the cNMP-binding domains are unique to PRKG1B . An example of a domain-only match is shown in Figure 2B . Neural cell adhesion molecule 2 ( NCAM2 ) and PDGFRB share two Ig domains , resulting in a significant alignment , also with an E-value of 2 . 4e−8 , and alignment coverage of 24% . However , the genes that encode them are not homologous and they perform different functions: NCAM2 is involved in cell-cell adhesion with no enzymatic function . The ability to distinguish multidomain homologs from unrelated pairs that share a domain is essential to genomic analysis . The evolutionary relationship between a and b in Figure 1 supports inferences about genome evolution , organization , and function . The same inferences would not necessarily be justified by the evolutionary relationship between b and c . For example , chromosomal regions enriched with homologous gene pairs are likely to be homologous themselves . In contrast , enrichment with homologous domains does not support the inference that a pair of chromosomal regions is homologous . Heuristics based on similarity and alignment coverage ( the fraction of the mean sequence length covered by the optimal local alignment ) have been proposed to screen out domain insertions . Recently , approaches based on domain architecture comparison have also been proposed [23]–[26] . To our knowledge , despite the prevalence of methods based on sequence similarity and alignment coverage [27]–[37] , the accuracy of these heuristics has never been systematically tested . However , the examples in Figure 2 raise doubt about the general effectiveness of these methods . Both pairs have weak sequence similarity , short alignments , and a similar combination of shared and unique domains . Setting a significance threshold to eliminate NCAM2 would also eliminate roughly 240 sequences that are related to PDGFRB , since more than a quarter of the Kinases that match PDGFRB have E-values less significant than 2 . 4e−8 . Alignment coverage would not help distinguish these two cases: the homologous pair has a shorter alignment than the unrelated pair . Nor could we separate this case by comparing domain content , since PDGFRB and PRKG1B share one domain , while PDGFRB and NCAM2 share two . For this example , sequence similarity , the length of the shared region , and domain architecture comparison all fail to distinguish the homologous pair from the domain-only match . To determine the extent of this problem , here we evaluate sequence similarity , alignment coverage , and domain architecture comparison on a hand-curated benchmark of 853 , 465 known homologous pairs . Our results show that these heuristics are all insufficient for consistent , reliable identification of multidomain homologs . Surprisingly , given its widespread use , even a modest alignment coverage requirement dramatically increased the number of mis-assigned homologs in our study . These results challenge two unstated , but widely accepted hypotheses: ( 1 ) homologous sequences share similarity along the bulk of their length and ( 2 ) the local alignment between homologous sequences usually covers a greater fraction of their mean length than the local alignments of sequences that only share a domain . These observations suggest to us that sequences alone may not consistently contain enough information to differentiate homology from domain-only matches . We introduce a novel method , called Neighborhood Correlation , that leverages additional information contained in the weighted sequence similarity network to distinguish homologs from domain-only matches . In this network , each vertex corresponds to a sequence . Vertices whose corresponding sequences have significant similarity are connected by an edge with weight proportional to that similarity . The neighborhood of a sequence is the set of vertices adjacent to it; that is , the set of all sequences that match it above a predefined significance threshold . ( In this work , “sequence neighborhood” refers to the local context of the sequence in the network and not to the region immediately surrounding it in the genome . ) Our analysis demonstrates that the neighborhood structure of gene pairs related through shared domain insertions is characteristically different from that of pairs related through duplication or speciation . These differences in neighborhood organization are detectable and can be exploited to distinguish homology from domain sharing . A homology detection method for genomic analysis must meet the following criteria: It should correctly predict homologous pairs and reject unrelated pairs , including those that share domains . With a single set of parameter values , it should perform reliably on sequences with a broad range of attributes , including single domain families , multidomain families , families with short regions of conservation , and families with weak sequence homology . Finally , it should be easy to use and fast enough for datasets comprising hundreds of millions of sequence pairs . In an empirical evaluation , we demonstrate that Neighborhood Correlation meets these criteria . It is highly effective in classifying multidomain homologs and achieves superior performance in comparisons with sequence similarity ( BLAST and PSI-BLAST ) , alignment coverage , and domain architecture comparison . To evaluate performance , we hand-curated a benchmark of 853 , 465 known homologous pairs of mouse and human sequences , drawn from twenty well-studied families . Our test set includes single-domain families , as well as multidomain families with promiscuous domains that are at risk for domain-only matches . Although comprehensive datasets are available for testing methods for predicting homology of individual domains [38] , [39] , we are unaware of any other gold-standard dataset of known multidomain families with variable domain architectures . We offer this validation dataset , which is based on published evidence by experts on each of the families , as a resource for future studies . As a validation of our approach , we applied Neighborhood Correlation to all complete , mouse and human sequences in SwissProt 50 . 9 to predict homologs . A comparison of our predictions with the euKaryotic Clusters of Orthologous Groups ( KOGs ) database [40] showed that the set of protein sequences with highly correlated neighborhoods includes the vast majority of pairs that share an orthologous group ( i . e . , have the same KOG annotation ) . This is consistent with the fact that orthology is a more restrictive criterion than homology . We also show that most pairs in our set of predictions share at least one domain , according to the Pfam database [41] , but many sequence pairs that share a domain are excluded . This is consistent with our goal of identifying gene homology rather than domain homology .
Here , we propose a model of multidomain homology based on vertical descent and insertion of a sequence fragment into an existing gene . In our model , two sequences are homologous if they are encoded by genes that share an ancestral locus . The rationale for this definition is illustrated in Figure 3 , which shows the evolution of genes through vertical descent and domain insertion in the context of the chromosomes in which they reside . When genomic context is taken into account , it is clear that genes g2 and g2′ are homologous , despite the fact that g2 contains a domain not present in g2′ and vice-versa . In contrast , genes g2 and g3′ are not homologous , despite the fact that they share a homologous domain , since g2 and g3′ are not located in chromosomal regions that share common ancestry . For comparative mapping applications , where homologous genes are used as markers for identifying chromosomal regions , this distinction is crucial . For example , phylogenetic footprinting [12] predicts transcription factor binding sites by identifying homologous genes and then searching their flanking chromosomal regions for conserved sequence motifs . In Figure 3 , the regions upstream of g2 and g2′ have an elevated probability of sharing conserved motifs since they share common ancestry . However , there is no reason to expect an enrichment of motifs shared between the flanking regions of g2 and g3′ . Our model is applicable to families that evolved through acquisition of a new domain by an existing gene . This can occur through insertion of sequence fragments into the gene or by recruitment of adjacent exons . Formation of a new gene architecture by domain loss is also consistent with our model . Several lines of evidence suggest that acquisition of an auxiliary domain by an existing gene is a relatively common mode of domain shuffling . First , a substantial number of metazoan , chordate , and vertebrate families have been identified that evolved through a pattern of duplication , insertion of domains , and further duplication , a pattern consistent with this model [46] , [47] . Second , the existence of promiscuous domains that lend themselves to insertion in new chromosomal environments [48] , [49] supports an insertion model . Third , domain insertion is more likely to be successful when a domain is inserted into an existing functional environment , e . g . , into the intron of an existing gene . In this case , all regulatory and termination signals required for successful transcription are already present . A fourth line of evidence stems from analyses of the flanking DNA of genes that arose very recently , where traces of the particular domain shuffling mechanism that occurred can still be observed . A number of recently evolved metazoan genes have been discovered that arose through duplication of an existing gene , followed by acquisition of one or more domains by unequal crossing over or by retrotransposition [50]–[54] . Finally , a number of studies have inferred relative rates of various domain shuffling events by applying parsimony models to abstract domain architectures . Their results suggest that the most common domain shuffling scenario involves insertion or deletion of a single domain into an existing multidomain architecture [24] , [55] , [56] . Our model is not applicable to the case where a new domain architecture is assembled de novo from several unrelated building blocks and subsequently acquires a regulatory region . We consider such a novel architecture to be the progenitor of a new family , since it is not clear that the ancestry of any one constituent is preferred . Similarly , our model does not capture formation of new architectures through fragmentation of more complex ones . However , recent evidence suggests that both of these scenarios occur rarely [24] , [55] , [57] . Homology detection is the problem of distinguishing between sequence pairs with different types of evolutionary histories: evolution via gene duplication or via domain insertion . Sequence similarity , alignment coverage , and domain architecture comparison have all been considered for this purpose . However , none of these distinguish the homologous pair from the domain-only match given in Figure 2 . The empirical results in the following sections confirm that this is not an isolated example . Accurate classification of multidomain homologs requires additional information from another source . The structure of the sequence similarity network provides a basis for distinguishing pairs related through vertical descent from other pairs . The local network neighborhoods of homologs and domain-only matches differ in both topology and edge weights . In particular , for homologous pairs , the shared neighborhood ( i . e . , the set of vertices adjacent to both members of the pair ) tends to have more vertices and stronger edge weights than their unique neighborhoods ( i . e . , vertices adjacent to one pair but not the other ) . This is not true for domain-only matches . We express this distinction quantitatively by the Neighborhood Correlation score of two sequences , defined to be the correlation coefficient of their respective neighborhoods: ( 1 ) where S ( x , i ) is the normalized bit score [58] of the optimal local alignment of query sequence x and database sequence i , N is the number of sequences in the database , and is the mean of S ( x , i ) over all sequences ( see Methods ) . Note that NC ( x , y ) increases with the number , weight , and correlation of edges in the shared neighborhoods of x and y and decreases with the number and weight of edges in their unique neighborhoods . The Neighborhood Correlation score captures properties of the sequence similarity network that are strongly influenced by the evolutionary processes of interest . The number of edges in the shared and unique neighborhoods is influenced by the rates of gene duplication and domain insertion , while edge weights depend on sequence divergence . Immediately following a gene duplication , the two resulting paralogs have identical neighborhoods . The Neighborhood Correlation score of this new pair is initially one and decreases as the sequences diverge . Additional gene duplications in the same family further increase the size of the shared neighborhood and , hence , the Neighborhood Correlation score . In contrast , if a domain is inserted into a single member of the pair , the number of edges in its unique neighborhood increases and the Neighborhood Correlation score decreases . The increase in the number of unique edges is directly related to the promiscuity of the inserted domain , while the weights of these new edges are proportional to the degree of sequence conservation in the domain superfamily . In practice , the impact of insertion of a domain into a single member on the Neighborhood Correlation score is typically small because promiscuity and sequence conservation within domain superfamilies are inversely related . For example , Pfam domains exhibit a highly significant , negative correlation between domain promiscuity ( see Methods ) and sequence identity ( ρ = −0 . 21 , p = 2 . 08e−30 , Spearman test ) . This can be understood by observing that when a domain is inserted into a new context , it is likely to experience new selective pressures leading to rapid mutational change . To see how these principles play out in practice , we consider the neighborhoods of PDGFRB , PRKG1B , and NCAM2 in the sequence similarity network derived from our test dataset ( Figures 2 and 4 ) . Although the homologous pair , PDGFRB and PRKG1B , and the domain sharers , PDGFRB and NCAM2 , have pairwise alignments with similar properties ( E-value , alignment length , number of shared domains ) , their neighborhoods in the weighted sequence similarity network are very different . The shared neighborhood of the Kinase homologs PDGFRB and PRKG1B is substantially larger ( 779 sequences ) than their unique neighborhoods ( 183 and 142 sequences , respectively ) . The shared neighborhood consists almost entirely of Kinases . The unique neighborhoods are dominated by domain-only matches , due to Ig in the case of PDGFRB and the cNMP-binding domain in the case of PRKG1B . Sequence similarities within these unique neighborhoods are weak; the Pfam models for the Ig and cNMP-binding domains have average sequence identities of 20% and 18% , respectively . Thus , the edge weights ( not shown ) in the shared neighborhood are strong and well correlated , while the edge weights in the unique neighborhoods are weak , yielding a Neighborhood Correlation score of NC = 0 . 65 . Conversely , PDGFRB and NCAM2 are related through domain insertion and have significant sequence similarity due to a shared Ig domain . Their shared neighborhood is relatively small ( 242 sequences ) and comprised primarily of Ig-based matches . These contribute little to the Neighborhood Correlation score of this pair due to low sequence conservation within the Ig superfamily . In contrast , the unique neighborhood of PDGFRB is large ( 630 sequences ) , with strong edge weights . For these reasons , PDGFRB and NCAM2 have a Neighborhood Correlation score of 0 . 29 , distinctly smaller than the score for PDGFRB and PRKG1B . Unlike sequence comparison , this clear difference in neighborhood structure can be used to recognize multidomain homology . Evaluation of classification performance requires a trusted set of positive examples ( known homologous pairs ) and negative examples ( pairs known not to share common ancestry ) . Although benchmarks are available for detection of remote homology ( e . g . , SCOP [38] , CATH [39] ) , functional similarity ( e . g . , the Gene Ontology ( GO ) [59] ) , orthology ( e . g , COGs [40] ) , and structural genomics ( [16] , 45 , [60] , and work cited therein ) , we are unaware of any gold-standard validation dataset for multidomain homology . Our benchmark is designed to be suitable for testing two classification goals: good overall performance on a large set of sequence pairs and consistent performance on individual families with varying properties . To satisfy these needs , we constructed a test set of 1577 sequences from 20 families of known evolutionary origin ( Table 1 ) . The families encompass a broad range of functional categories , summarized in Table 2 . The full curation procedure is described in Methods and Text S1 . For each family , we identified two sets of sequence pairs: family ( FF ) pairs , where both members of the pair are in the family , and non-family ( FO ) pairs , where only one of the two sequences is in the family . Given a family of size k , we obtain k2 FF pairs ( the positive examples ) and k ( N−k ) FO pairs ( the negative examples ) . Individual families , which cover a range of functional properties and domain architecture complexity , can be used for family specific tests . In addition , we constructed a test set ( ALL ) for general performance evaluation by merging all sets of FF and , respectively , FO pairs , yielding 853 , 465 positive and 40 , 459 , 204 negative examples . Performance measurements obtained with this set could be biased by the Kinase family , which is much larger than the other families . We therefore also considered the set of all sequences excluding the Kinases ( ALL-Kin ) , resulting in 32 , 629 positive and 17 , 545 , 558 negative examples . Our goal is a method that can correctly identify homologs in multidomain families without degrading performance in other types of families . We therefore devised a benchmark to test a range of homology detection challenges , involving single domain as well as multidomain families . Families with complex and varied domain architectures represent the primary challenge undertaken in this study . Such families result from duplication , domain accretion , and further duplication . Some of these families are defined by a single domain that is unique to the family ( e . g . , Kinase ) , while others are characterized by a particular combination of domains ( e . g . , ADAM ) or by a conserved set of domains with variations in domain copy number ( e . g . Laminin ) . Modularity in both single and multidomain families can also arise through the presence of sequence motifs , such as subcellular localization signals , transactivation sequences ( e . g . , Tbox ) , and functional components that confer substrate specificity ( e . g . USP ) . These motifs can result in matches to unrelated sequences . In addition , promiscuous domains challenge homology identification because they can result in significant sequence similarity but carry little information about gene homology . Promiscuity can confound reliable detection of homologs even in families with conserved domain architectures . Remote homology detection is a serious challenge that has received widespread attention . In our dataset , this challenge is represented set by FGF , TNF , TNFR , and USP , families that exhibit low sequence conservation . Finally , we considered homologous pairs with short conserved regions . A minimum alignment coverage criterion is frequently imposed to eliminate domain-only matches , reflecting a widely held , but untested belief that homologous pairs have regions of similarity that cover a substantial fraction of their length . To test the robustness of homology detection methods with respect to alignment length , we included single domain families with short conserved regions such as the Tbox family . Our selection of test families was limited to those for which it was possible to obtain evidence concerning their evolutionary history . Evolutionary evidence was obtained from published articles and/or curation by a nomenclature committee . In the best cases , direct syntenic evidence of vertical descent can be found . In other cases , indirect evidence such as conserved intron/exon structure is used . Phylogenetic evidence can confirm vertical descent , for example , if all domains in a family have consistent phylogenies . However , phylogenetic disagreement between core and auxiliary domains does not rule out homology according to our model . For each , the evidence used is described in Text S1 . We evaluated Neighborhood Correlation using our benchmark , and compared its performance with other methods currently in use . We considered performance on multidomain homology identification , as well as overall performance on diverse , heterogeneous datasets . We also used Neighborhood Correlation to predict novel homologous relationships . We compared the performance of Neighborhood Correlation with BLAST [61] , alignment coverage [27] , and PSI-BLAST [58] , methods commonly used for assessing homology , as well as Domain Architecture Comparison ( DAC ) , a recently introduced approach that compares sequences by considering their constituent domains [23]–[26] , [55] . BLAST gives a measure of sequence similarity based on the optimal local alignment between two sequences . BLAST does not capture gene structure ( e . g . , domain organization ) , nor does it reflect additional information that might be derived from suboptimal local alignments . BLAST is widely used , its behavior is well understood , and its scores are easily compared with those from other studies . A great deal of attention has been devoted to tuning BLAST performance and to developing accurate statistical tests . It represents an attractive balance between rigor and speed . A significant BLAST score is evidence of similarity greater than that expected by chance , but cannot distinguish whether that similarity stems from vertical descent or domain insertion . In order to eliminate domain-only matches , many analyses combine sequence similarity with alignment coverage to identify homologs [28]–[37] . To be considered homologous , sequence pairs must then satisfy a second criterion in addition to significant sequence similarity: the fraction of the sequence length covered by the optimal local alignment must meet a pre-specified threshold . To our knowledge , alignment coverage criteria have never been empirically evaluated . In this work , we demonstrate that such a requirement is highly detrimental to performance overall , and in nearly all tested families . In the presence of high sequence divergence , BLAST is limited by the amount of information that can be derived from pairwise comparison . To address this problem , approaches based on multiple sequence alignments ( MSAs ) have been used to increase sensitivity . PSI-BLAST , one of the most widely used examples of this approach , constructs a Position Specific Scoring Matrix ( PSSM ) through iterative search and has been shown to dramatically improve sensitivity [62] . MSA-based methods are designed to detect remote homology , not multidomain homology . Since sequences with different architectures cannot be aligned , MSA-based methods are not a natural choice for multidomain homology detection . We included PSI-BLAST in our study because it is widely used as a standard for remote homology detection . In addition to sequence based methods , we considered direct comparison of domain architectures for multidomain homology detection . Each sequence was represented by a linear sequence of Pfam domains . Linker sequences between domains were ignored , as was sequence variation between instances of a given Pfam domain family . The resulting domain architectures were compared based on their domain composition . In a previous study , we proposed and evaluated 21 different methods for comparing domain architectures [23] . These methods considered properties such as the number of shared domains , domain copy number , total number of domains in a protein , domain order , and domain promiscuity . We included the domain architecture comparison strategy that exhibited the best performance from that study in our current study . This method assigns a score to each pair based on the number of shared domains ( see Methods ) , following the rationale that homologous pairs will have more domains in common than pairs related through domain insertion . In assessing similarity , each domain is assigned a weight inversely proportional to its promiscuity . This reflects the assumption that rare domains convey more information about homology than promiscuous domains . The performance of each method was assessed via the ROC-n score ( Table 3 ) , which represents both false positives and false negatives ( see Methods ) . ROC-n is the area under the Receiver Operating Characteristic ( ROC ) curve comprised of the top ranking pairs up to the first n false positives . We used n = 100k , where k is family size , corresponding to 100 false positives per query . In evaluating homology identification methods , we consider two user models . Genome-scale analyses require all-against-all comparison of a large and heterogeneous set of sequences . In order to be suitable for automated , genomic analyses , a method must be robust enough for use without human intervention , deliver consistent behavior on different types of domain architectures , and be fast and easy to use . In this case , the goal is to maximize the total number of homolog pairs that are correctly predicted . A second application is analysis of individual families , where the goal is to obtain good per-family prediction scores over a wide range of families . To evaluate performance for both user models , we report ROC-100k scores for all pairs ( ALL and ALL-Kin ) , as well as ROC-100k scores for each family . To show how the methods tested behave on proteins with various attributes , we also report the average ROC-100k score per family for single domain families , multidomain families with conserved architectures , and multidomain families with variable architectures . As a visualization tool , we generated rank plots , which show the scores of all matches to a given query sequence in rank order . Rank plots provide a visual representation of the organizational structure of the network neighborhood of the query sequence , as well as organizational substructure within the family . For example , Figure 5 shows a rank plot for the query sequence PDGFRB , a protein tyrosine kinase . The break in the curve in Figure 5B at NC≈0 . 8 corresponds to the first match to a Serine/Threonine Kinase , the inflection point at NC≈0 . 75 corresponds to the first match to a Dual-Specificity Kinase , and the downward plunge at NC≈0 . 59 corresponds to the first Casein Kinase . Rank plots for each of the 26 , 197 sequences in our dataset are provided at http://www . neighborhoodcorrelation . org . When all considered classifiers are applied to the aggregate set of sequence pairs ( ALL ) , Neighborhood Correlation dramatically outperforms the other three methods ( Table 3 , Figures S1 and S2 ) . In the ALL-Kin dataset , Neighborhood Correlation yields better performance than BLAST and PSI-BLAST , but performs slightly worse than DAC . The superior performance of Neighborhood Correlation on the ALL and ALL-Kin datasets demonstrates that its optimal classification threshold is less sensitive to family specific properties than those of BLAST or PSI-BLAST . When performance on individual families is considered , Neighborhood Correlation is generally more robust than the other three methods . It perfectly classifies twelve families , more than any other method . In addition , in 16 of 20 families , the discriminatory performance of Neighborhood Correlation is better than or equal to that of all other methods . In particular , Neighborhood Correlation obtains the highest average score for both conserved and variable architectures and performs much better on individual multidomain families except for Myosin and Kinesin . For families with high sequence divergence , including FGF , TNF , and USP , Neighborhood Correlation performs better than BLAST , indicating that neighborhood structure can compensate for a low signal to noise ratio in pairwise comparisons of remote homologs . PSI-BLAST also performs well in such cases . To demonstrate why Neighborhood Correlation is more effective for complex families , we consider its performance on the Kinase family . Figure 5 shows a rank plot of the results of a query with the Kinase PDGFRB . A robust method is expected to rank all Kinase family members before non-Kinase matches . In particular , we examine pairing between the Kinase PRKG1B and the non-Kinase NCAM2 , the genes depicted in Figure 2 . Neighborhood Correlation exhibits no difficulty separating these pairs . The match with PRKG1B scores substantially higher than NCAM2 ( indicated by magenta and green circles , repectively , in Figure 5 ) . In contrast , the BLAST scores for these sequences are indistinguishable , and the PSI-BLAST scores for these sequences are reversed: The match to NCAM2 obtains θ = 3 . 65e−40 , while the match to PRKG1B is much less significant ( θ = 1 . 26e−25 ) . How typical are these examples ? As shown in Figure 6 , the sequence similarity distributions of FF and FO pairs overlap completely for BLAST and partially for PSI-BLAST . In contrast , the Neighborhood Correlation score distributions for family and non-family matches are largely distinct , with only a limited overlap in the tails of the distributions . Neighborhood Correlation also delivers robust performance when sensitivity ( Sn ) and specificity ( Sp ) are considered independently . For example , when matches to the query sequence PDGFRB are ranked by Neighborhood Correlation score ( Figure 5A ) , a cutoff of NC = 0 . 3 results in three false positives with only ten false negatives . In contrast , a BLAST threshold of E<3e−10 results in three false positives and 630 false negatives ( Figure 5B ) . The number of false negatives obtained with PSI-BLAST at this specificity is even greater ( Figure 5C ) . More generally , the ROC-n curves for the Kinase family in Figure 7 demonstrate that Neighborhood Correlation achieves both higher sensitivity and higher specificity than BLAST , except at very high specificity , and always outperforms PSI-BLAST by both measures . Neighborhood Correlation simultaneously achieves Sn≈0 . 85 and Sp≥0 . 999 . At this specificity , Sn≈0 . 7 for PSI-BLAST and Sn≈0 . 55 for BLAST . While the other methods considered have strengths specific to particular challenges , Neighborhood Correlation delivers the most reliable and consistent performance on large , heterogeneous datasets . Neighborhood Correlation is , therefore , particularly well suited to automated genome-scale analyses , which require that a single classification threshold be suitable for the vast majority of sequence pairs in a genomic dataset . Moreover , Neighborhood Correlation is robust . The distribution of Neighborhood Correlation scores for all sequence pairs in our dataset ( Figure S3 ) has a flat trough ranging from 0 . 4 to 0 . 8 . Within this range , the prediction quality will be relatively insensitive to the choice of threshold . A putative set of mouse and human homologs imposed by a threshold of NC≥0 . 6 on all sequence pairs in our dataset is available at http://www . neighborhoodcorrelation . org . As expected , PSI-BLAST excels at families with low sequence conservation , such as TNF and USP , and generally performs well on single domain families . However , PSI-BLAST falters on complex multidomain families and on sequences with promiscuous domains . PSI-BLAST's average ROC-100k scores for both conserved and variable multidomain families are inferior to those of both Neighborhood Correlation and BLAST . This is exemplified by PSI-BLAST's poor performance ( Figure 5B ) when querying with PDGFRB , which has two copies of the highly promiscuous Ig domain . PSI-BLAST's iterative profile construction algorithm incorporates matches to the highly promiscuous Ig domain in the growing alignment , even when a very stringent inclusion threshold ( E<10−13 ) is used . As a result , unrelated sequences that contain Ig domains match the resulting profile with better scores than Kinases without Ig . PSI-BLAST performs better on the Kinase family as a whole than it does on PDGFRB ( Table 3 ) because many Kinases are single domain proteins . When classification of heterogeneous data is considered , PSI-BLAST's performance is inferior to Neighborhood Correlation on the ALL dataset and to both Neighborhood Correlation and BLAST on the ALL-Kin dataset . This demonstrates that no single PSI-BLAST cutoff is suitable for all families . Indeed , inspection of PSI-BLAST output on individual queries ( data not shown ) indicates that PSI-BLAST scores tend to vary widely from family to family . PSI-BLAST introduces a clear tradeoff between sensitivity and generality , to the particular detriment of large-scale studies . Moreover , PSI-BLAST is characterized by greater instability and running time than BLAST or Neighborhood Correlation . Domain architecture comparison performs well on single domain families and on multidomain families with conserved domain architectures ( e . g . , DVL , Notch , Laminin , and WNT ) . Like PSI-BLAST , DAC can recognize distant homology because domain architectures are recognized by MSA-based models . The performance of DAC on other families is mixed , however , because it faces a number of challenges that do not arise with the other classification methods . First , all domain architecture comparison methods are substantially restricted by the limitations of domain detection . In our dataset , 12 . 7% of sequences do not have domain annotations , resulting in low ROC-100k scores for many families . This explains why single domain families , such as Tbox , which have identical domain architectures , do not achieve perfect ROC-100k scores , contrary to expectations . An additional shortcoming is that domain architecture comparison methods do not capture information in linker sequences or sequence variation within a domain family . Therefore , domain architecture comparison tends to assign the same score to pairs that actually differ in sequence divergence . This explains the long plateaus in the ROC curve for DAC in Figure 7 . A particularly challenging problem for domain architecture comparison is how to effectively distinguish domains that proliferated through gene duplication from promiscuous domains that proliferated through domain shuffling . The number of domain partners , used here , is a typical measure of promiscuity , based on the assumption that this measure reflects the frequency of domain insertion [48] . This measure of promiscuity will inappropriately down-weight a domain that characterizes a family , if the domain happens to be the target of insertions of many other domains . Consider , for example , a sequence with a single domain A that sustains repeated duplication , followed by insertion of different domains into the resulting copies , yielding AB , AC , AD , and so on . Domain A will have a high promiscuity score , although it is never inserted into new contexts . As a concrete example , the Pkinase domain partners with more than 100 different domains . However , the resulting high promiscuity score may be inappropriate since Pkinase lacks many of the other characteristics of promiscuous domains , such as small size and 1-1 phase [17] , and is important in defining the Kinase family . This explains why domain architecture comparison performs poorly on the Kinase family . To assess the effectiveness of alignment coverage in eliminating domain-only matches , we compared ROC-100k scores for sequence similarity alone and combined with alignment coverage ( α , see Methods ) . We considered three alignment coverage thresholds , α≥0 . 3 , α≥0 . 6 , and α≥0 . 8 , that span the range of length cutoffs used in the literature ( e . g . [32] , [34] ) . The results ( Table 4 ) show that the addition of an alignment coverage criterion does not improve the performance of sequence similarity . For example , a cutoff of α≥0 . 3 reduces the ROC-100k score by 25% in the ALL dataset and 23% in the ALL-Kin dataset . When families are considered individually , a cutoff of α≥0 . 3 decreases the ROC-100k score by at least 10% in one-third of the families . Increasing the cutoff to α≥0 . 6 or α≥0 . 8 does not increase performance in any family . Note that although the ROC-100k score for KIR when α≥0 . 6 is higher than the score for sequence similarity alone , this difference is not significant ( p = 0 . 69 ) . Alignment coverage is based on the assumption that non-homologous pairs have shorter regions of similarity than homologous pairs , yet Table 4 suggests this is not universally true . To assess the extent to which the region of similarity in homologous pairs extends over the bulk of their length , we calculated Precision and Recall ( see Methods ) for α≥0 . 3 , α≥0 . 6 , and α≥0 . 8 . The results , shown in Tables 5 and Table S1 , suggest that full length alignments are not a characteristic property of homologous families , at least in our dataset . In the ALL-Kin dataset , a cutoff of α≥0 . 3 eliminates 40% of true positives , specifying α≥0 . 6 eliminates 70% of true positives , and α≥0 . 8 eliminates 83% true positives . The loss in Recall is even more extreme in the ALL dataset . To better understand these results , we plotted histograms of α for individual families ( Figures 8 , S4 ) . While some families do have long regions of similarity , long conserved regions are not a persistent characteristic of most families in our dataset . Several different trends in domain organization can cause this . Some families are characterized by a short , conserved domain , such as the DNA binding domain in the FOX family , and little conservation elsewhere ( Figure 8A ) . Multidomain families exhibit a range of alignment lengths for a variety of reasons . In families characterized by a single defining domain partnered with a variety of auxiliary domains , alignment lengths depend upon the number of domains a given pair has in common . For example , the histogram for the PDE family ( Figure 8B ) has a small peak near α = 1 . 0 , corresponding to pairs with identical domain architectures , and a much larger peak between α = 0 . 2 and α = 0 . 7 that represents pairs of family members with different auxiliary domains . Families can also demonstrate wide variation in due to differences in copy number ( e . g . , Laminin , Figure 8C ) . Finally , a broad α distribution can be caused by variation in sequence length within the family . Even when the length of the conserved region is constant , alignment coverage , expressed as a fraction of total length , may vary widely , confounding homology prediction methods based upon alignment coverage . Given the widespread use of alignment coverage criteria , we were surprised by this poor performance . We examined the possibility that our failure to observe a consistent pattern of long alignments was due to the fact that we considered the length of the optimal alignment , only . To investigate whether including sub-optimal alignments would result in different conclusions , we implemented a simple heuristic ( see Methods ) that identifies and combines a consistent set of high-scoring local alignments; i . e . , alignments that appear in the same order in both sequences and do not overlap . Surprisingly , including suboptimal alignments in the alignment coverage calculation has little impact on our results . The distributions of the combined alignment lengths , shown in turquoise and brown in Figures 8 and S4 , differ little from the distribution of optimal alignment length distributions ( shown in blue and red ) . Nor do the values of Precision and Recall obtained with combined alignments differ greatly from those obtained with the optimal alignment ( see Table 5 and Table S2 ) . In summary , analysis with combined alignments confirms that full length similarity is not a general characteristic of homologous families .
Although models of gene family evolution have been proposed and debated for more than three decades [18] , models of multidomain evolution are in their infancy . Gene homology is a yes/no question: genes either share common ancestry or they do not . With this in mind , Fitch [42] argued that when subsequences of genes have distinct evolutionary histories , it is not possible to determine gene homology . Rost and colleagues [45] , [63] further proposed that “dissecting proteins into structural domain-like fragments” [45] is the only reasonable way to study relationships in such proteins . We suggest an alternative: By considering the genomic context of genes that encode multidomain proteins , it is possible to define homology for multidomain sequences without violating the tenet that homology is an indivisible property . We propose a model of multidomain evolution in which the set of events by which sequences diverge is expanded to include domain insertion and deletion as well as mutation . Recent evidence from studies of young genes [50]–[53] , as well as indirect evidence of sequence shuffling [17] , [24] , [49] , [55] , [56] , suggests that our model is consistent with a significant fraction of metazoan multidomain families . This model permits discrimination between genes related by vertical descent and those related by domain insertion alone , which is the basis for our definition of multidomain homology . This in turn enlarges the scope of inquiry from domain family homology to gene family homology , providing a broader context in which to study the evolutionary processes by which modular families are formed . Our model does not describe families that evolved through other domain shuffling processes such as gene fission , the fusion of adjacent genes resulting from read-through errors , or de novo formation of novel architectures through independent insertions in intergenic regions . Extending the model to capture a broader range of domain shuffling scenarios and testing it on other datasets and applications are important directions for future work . Evidence supporting the validity of our model can be obtained by comparing Neighborhood Correlation with related classifications , such as orthology and domain homology . The success of Neighborhood Correlation in recapitulating homologous relationships in our benchmark empirically supports Neighborhood Correlation as a predictor of homologous genes; that is , sequences derived from a common ancestor by vertical descent , whether by duplication or speciation . Since orthologs , sequences that diverged by speciation in their most recent common ancestor , are by definition homologs , our model predicts that known mouse and human orthologs will have high Neighborhood Correlation scores . To test this prediction , we compared Neighborhood Correlation with KOGs [40] . As expected , 90% of sequences in our dataset with the same KOG annotation have a Neighborhood Correlation Score greater than 0 . 6 ( Figure 9A ) . However , only 12% of pairs with NC≥0 . 6 share the same KOG annotation . This is consistent with the observation that gene homology is a necessary but not sufficient condition to establish orthology . Domain homology , on the other hand , is a less stringent criterion than gene homology . Homologous genes , by definition , share at least one homologous domain . Of pairs with Neighborhood Correlation scores above 0 . 6 , 88% of pairs share at least one Pfam [41] code ( Figure 9B ) , consistent with the assertion that gene homology is a more stringent requirement than domain homology . That the remaining 12% do not share a domain is primarily due to missing annotations . Recall that 12 . 7% of sequences in our dataset do not contain a recognizable Pfam domain . Since only some sequences that share a domain are encoded by homologous genes , our model predicts that a significant fraction of sequence pairs that share homologous domains will not have high Neighborhood Correlation scores . In fact , with NC≥0 . 6 , only half of sequence pairs in our dataset share a Pfam domain . These results are consistent with the expectation that gene homology is a less restrictive condition than orthology but more restrictive than domain homology . This analysis provides additional evidence , independent of our curated dataset , that Neighborhood Correlation can predict homologous genes according to our model . Insight into the evolutionary processes responsible for the development of novel function are of greatest value when considered in the context of entire genomes . To accommodate studies of such scale , a method must be suitable for robust , automated analyses . For the current application , this requires speed , ease of use , and consistent behavior across varied domain architectures . Neighborhood Correlation displays excellent performance across an array of families with a range of sequence patterns and evolutionary histories . Neighborhood Correlation is able to correctly classify complex families , while maintaining accuracy on simpler families . Further , it displays a classification threshold that is robust with respect to family , yielding good performance on individual families as well as on aggregate datasets in which families may not be known or readily discernible . Since Neighborhood Correlation can be computed easily with existing computing resources and data stores , it is easy to add to a computational workflow . These qualities demonstrate that Neighborhood Correlation is well suited to large-scale genomic analysis . Empirical evaluation of existing homology detection methods revealed limitations in their applicability , often contrary to common expectations . Meticulous tests of BLAST and PSI-BLAST performance have been carried out on well-characterized datasets [58] , [62] , [64] , but , to our knowledge , performance on multidomain proteins with promiscuous domains and low complexity regions has not been considered empirically . Our tests on datasets with multidomain sequences , promiscuous domains , and low complexity regions show that while BLAST represents an attractive balance between speed and accuracy on conserved , single-domain families , additional screening is needed for correct multidomain classification . Since Huynen and Bork [27] proposed that alignment length could be used to reduce false positives in ortholog prediction , the practice of pre-screening using an alignment coverage criterion has become widespread in genomic analyses [28]–[37] . To determine the effectiveness of this approach , we investigated the two hypotheses underlying the use of alignment coverage: Surprisingly , the imposition of an alignment coverage requirement , in addition to sequence similarity , actually decreased the accuracy of homology identification , suggesting that the above hypotheses are not generally true . To our knowledge , this is the first rigorous evaluation of alignment coverage . Our study suggests that PSI-BLAST , while first-rate for detecting remote homology , is ill-suited to large scale automated analyses on datasets with complex multidomain architectures , promiscuous domains , and low complexity sequences due to its running time , instability , and family dependent score thresholds . The same iterative strategy that confers PSI-BLAST's increased sensitivity leads to a lack of robust behavior when PSI-BLAST is run in an automated manner . Even at extremely stringent inclusion thresholds , false positives are incorporated in during model construction when the query sequence contains promiscuous domains or low complexity regions . Once a false positive is included , PSI-BLAST rapidly degrades the MSA used in subsequent iterations , leading to both incorrect results and excessively long running times . PSI-BLAST required 208 CPU days for our dataset , a 300-fold increase in time over basic BLAST . This slowdown is associated with the large fraction of promiscuous , multidomain , and low complexity sequences in our dataset . When PSI-BLAST is used interactively , the user can eliminate potentially troublesome matches by inspection; however , human intervention is not possible for genome-scale studies . The additional computational cost of calculating Neighborhood Correlation scores once a BLAST search has been performed is negligible . Though PSI-BLAST does offer accuracy improvement over Neighborhood Correlation on families with conserved domain architectures , these issues suggest that PSI-BLAST is impractical for this or larger genomic studies . Domain architecture comparison performs well on families with low sequence conservation due to the discrimintatory power of multiple alignment based domain models , yet our empirical evaluation of DAC reveals several areas for improvement . Domain architecture comparison can be compromised by faulty or incomplete domain annotation . Failure to capture sequence variation within domain and linker sequences results in an inability to resolve family substructure . A model of promiscuity that better captures domain mobility is needed to correctly classify families defined by a single domain with many partners . Because the sequence similarity network reflects both domain architecture and sequence variation , Neighborhood Correlation avoids many of these difficulties , including unresolved family substructure and sensitivity to domain annotation . Neighborhood Correlation captures modular organization on a range of scales , including sequence motifs as well as structural domains , regardless of whether these subunits are encoded in a database . In addition , Neighborhood Correlation's success on kinase classification , relative to DAC , suggests that it may be possible to derive accurate promiscuity measures from the network . Neighborhood Correlation differs fundamentally in both goals and approach from Position Specific Scoring Matrices , Profile hidden Markov models , PSI-BLAST , and similar methods that exploit multiple alignments to detect distant homology . MSA-based approaches are not suitable for detecting multidomain homologs with varied architectures . These rely upon full length alignments that are not possible with multidomain sequences . The objective of multiple alignment methods is to identify related sequence motifs when the signal to noise ratio is low . In contrast , the goal of Neighborhood Correlation is to identify homologs that have sustained domain insertions and deletions since their divergence . Neighborhood Correlation also differs from methods based on multiple alignment in its computational approach . Although both approaches derive information from neighboring sequences , only Neighborhood Correlation exploits the topology of the network . MSA-based methods synthesize a model from a set of neighbors in the sequence similarity network and then use the resulting composite model in pairwise comparisons . Such models reflect aggregate properties of the network neighborhood , but not the underlying topological structure of the network . In contrast , Neighborhood Correlation compares the edge weights for each pair of shared neighbors separately , capturing not only neighborhood membership , but also specific information about how individual sequences in the neighborhood are related . Finally , Neighborhood Correlation derives information from neighborhood difference as well as from neighborhood similarity , taking advantage of the fact that sequences that match one member of the pair and not the other are informative . Neighborhood Correlation complements a recent set of studies relating multidomain evolution to the global topological properties of the domain similarity network [65]–[69] . Unlike these methods we focus on local network structure as evidence of the evolutionary history of specific sequence pairs and families . In an early use of local network structure , Koonin and colleagues [40] argued that orthologous groups correspond to cliques in the sequence similarity network . In a similar vein , Przytycka and colleagues [70] , [71] used a different aspect of local structure ( chordality ) to test whether domain insertion and intron acquisition are evolving in a parsimonious manner in a given family . In a recent study of protein families in prokaryotes , Medini et al . [72] consider local network structure , but do not relate it to evolutionary processes . In their study , they developed a scoring system based on sets of nearest neighbors in an unweighted network and used these pairwise scores to identify core sets of proteins associated with secretion systems in prokaryotes . Neighborhood Correlation links local network structure to both domain architecture and evolutionary process . The similarities and differences in domain architecture are reflected in the neighborhoods of adjacent sequences . The number and weights of edges in the shared neighborhood is influenced by the number and conservation of their shared domains . Their unique neighborhoods are similarly influenced by their unique domains . The Neighborhood Correlation score , therefore , is an implicit measure of both sequence similarity and domain architecture comparison . The history of gene duplication and domain insertion in gene family evolution is also recorded in network topology . Neighborhood Correlation is able to elucidate multidomain homology because it can decipher the traces of this history in the network . In particular , Neighborhood Correlation relies on the hypothesis that the neighborhoods of genes related through duplication are more similar to each other than the neighborhoods of genes related through domain insertion . This hypothesis in turn assumes that There is concrete evidence to support the latter assertion as indicated by the negative correlation between the promiscuity and sequence identity of Pfam domains , discussed in Results . We are not aware of any studies predicting the relative rates of gene duplication and domain insertion . However , the success of Neighborhood Correlation in classifying multidomain homologs provides indirect evidence that the assertion is true , at least in the dataset studied here . If , contrary to this hypothesis , domain insertions occurred as or more frequently than gene duplications , the Neighborhood Correlation scores of multidomain homologs would not be distinctly higher than those of domain-only matches . More generally , the success of Neighborhood Correlation has demonstrated that information about the interplay of the processes of gene duplication , domain shuffling , and sequence divergence lies hidden in the local structure of the sequence similarity network . This success suggests that mining network structures is a promising direction for extending bioinformatics methodology , as well as for asking basic questions about evolutionary processes . For example , it has been argued that the increased complexity of multidomain families in metazoans is directly related to the advent of multicellular animals . Multicellularity has evolved several times ( [73] and work cited therein ) . In each case , Nature has had to evolve novel solutions to the problems of coordinated cellular communication and control . It is an intriguing question whether the same patterns of gene duplication and domain insertion that prompted the evolution of metazoan signal transduction families also dominate in other lineages . Future work will determine whether we can further exploit local organization of the sequence similarity network to investigate such questions .
We extracted all complete mouse and human protein sequences from SwissProt Version 50 . 9 [74] , yielding 11 , 553 mouse protein sequences and 14 , 644 human protein sequences . Sequence fragments were excluded from this set of sequences by rejecting sequences annotated with a description field containing “ ( fragment” . We chose SwissProt , a high quality , curated protein sequence database , as opposed to GenBank , which would have resulted in a larger , but less reliable , dataset . KOG annotations were obtained from the Clusters of Orthologous Groups database [40] , available from ftp://ftp . ncbi . nih . gov/pub/COG/KOG/ . KOG annotations were mapped to SwissProt identifiers by exact matching of KOG FASTA protein sequences with those in SwissProt . The analysis was carried out on the combined set of mouse and human sequences . In a preliminary study , we compared the performance of Neighborhood Correlation on a smaller , combined set of mouse and human sequences with its performance on separate sets of mouse and human sequences [75] to determine whether Neighborhood Correlation performs differently on comparisons within and across genomes . The mouse-only and human-only data test the ability to classify paralogs within a single mammalian species , as opposed to the combination of orthologs and paralogs seen in the combined dataset . The basic trends in the mouse-only and human-only datasets were the same as the combined dataset for all tests performed . This suggests that Neighborhood Correlation performance is not highly sensitive to the degree of sequence divergence , since paralogous and orthologous sequences in these species exhibit different patterns of divergence . For each family , we derived a list of designated gene symbols , Pfam [41] and/or InterPro [76] codes from publications by family experts , and reports from the Human Genome Nomenclature Committee ( http://www . gene . ucl . ac . uk/nomenclature/genefamily . html ) . These lists were used to generate a preliminary roster for each family , then confirmed by referring to recent analyses of gene family evolution in the literature . A detailed account of the curation procedure for each family with specific identification criteria and references is given in Text S1 . SwissProt accession numbers for all sequences in the twenty families are provided in Dataset S1 . We conducted all-against-all BLAST ( Version 2 . 2 . 15 ) [61] and PSI-BLAST ( Version 2 . 2 . 16 ) [58] searches for the sequences in our dataset , using the BLOSUM 62 matrix , an affine gap penalty of − ( 11+k ) for a gap of length k , and low complexity filtering . For both searches , the size of the search space was set to Y = n2 and the significance threshold to E = 10N , where n is the size of the database in residues and N is the number of sequences in the dataset . The combined dataset has N = 26 , 197 sequences , 11 , 553 mouse and 14 , 644 human sequences , corresponding to a total of n = 14 , 073 , 417 residues . For PSI-BLAST , four passes were executed with an inclusion threshold of E<10−13 for inclusion in the multiple alignment used to search in the next pass . Although this cutoff is much more stringent than the default , we found it essential to obtain correct results with sequences containing low complexity regions . Less stringent thresholds resulted in the inclusion of unrelated sequences in the intermediate PSSM . Asymmetries ( i . e . , E ( x , y ) ≠E ( y , x ) ) that occur due to low complexity filtering [77] , which is applied only to the query sequence but not to database sequences , were corrected by assigning the better of the two values to both matrix entries . The resulting dataset had 4 , 864 , 226 significant BLAST pairs and 10 , 854 , 626 significant PSI-BLAST pairs . The parameter values used in this study embody the view that an all-against-all BLAST search is a single experiment . This approach is roughly equivalent to conducting N single query BLAST searches with E = 10 and Y = mx n , where mx is the length of query sequence x . Treating the all-against-all BLAST comparison as a single experiment results in symmetric E-values in the absence of low complexity filtering . We define θ ( x , y ) = E ( x , y ) /10N to be the expected number of chance hits per sequence in the dataset with a score equivalent to , or better than , that of the alignment of query sequence x with matching sequence y . The significance threshold of E = 10N corresponds to θ = 10 chance hits per sequence , in expectation . We calculated the Neighborhood Correlation scores for all sequence pairs in our dataset from Equation 1 using the similarity score , ( 2 ) where ς ( x , i ) is the normalized bit score [58] of the alignment of x and i and ςmin ( x , i ) = log2 ( n2/10N ) *0 . 95 = 28 . 019 , which is 5% less than the bit score corresponding to θ = 10 for a dataset of the size used in this study . The effectiveness of Neighborhood Correlation depends strongly on how the similarity score , S ( x , i ) , is defined . We considered three measures of similarity: S ( x , i ) = log ς ( x , i ) , S ( x , i ) = ς ( x , i ) and an unweighted comparison of neighborhood membership defined as S ( x , i ) = 1 if there is a significant match between x and i , and zero otherwise . Although the other two measures performed well on some families , only S ( x , i ) = log ς ( x , i ) gave consistent , good performance on a wide range of families . This suggests two factors that may be important to Neighborhood Correlation performance . First , the relatively poor performance of the unweighted score indicates that it is necessary to capture differences in the degree of similarity to sequences in the neighborhood to capture complete evolutionary information . Second , the improved performance obtained with S ( x , i ) = log ς ( x , i ) can be understood by recalling that the correlation coefficient captures only linear associations . The use of the logarithm compresses the range of ς ( . , . ) , resulting in scores that more closely approximate linearity . The choice of ςmin , the score assigned to pairs without significant similarity , may influence Neighborhood Correlation performance in homology identification . We experimented with values of ςmin corresponding to significance thresholds ranging over two orders of magnitude . The results ( data not shown ) suggest that varying ςmin has little impact on Neighborhood Correlation . Promiscuity refers to the tendency of domains to be inserted into many different contexts . Typically , promiscuity of a domain is defined as the number of distinct partners associated with it , where two domains are partners if they co-occur in at least one sequence [3] . We obtained the set of Pfam codes associated with all sequences in our dataset from the SwissProt database . For each Pfam domain , we determined the number of distinct Pfam codes that co-occur with it in any of the 26 , 197 sequences in our dataset . We further obtained percent sequence identity for each Pfam identifier from the Pfam website . The Spearman ranked correlation coefficient of domain promiscuity and sequence identity was calculated to evaluate whether promiscuity and sequence identity were related . We conducted an all-against-all domain architecture comparison using the Pfam identifiers provided by SwissProt . Similarity of each pair of sequences , x and y were calculated as follows: ( 3 ) where w ( di , x ) is the weight of domain di in sequence x . Domains are assigned weights inversely proportional to their promiscuity . Promiscuous domains may occur in many unrelated sequences , and so are less useful than relatively rare domains in determining homology . The weight of a domain not contained in a given sequence is zero . As a result , pairs of sequences which share no domains are assigned a similarity of zero . This domain architecture comparison function corrects for the bias of proteins with many domains . Proteins with numerous domains have an elevated probability of sharing a domain with other proteins . Of the 21 domain architecture comparison methods we evaluated in a previous study [23] , this was shown to have the best performance . For every pair of sequences , x and y , with significant similarity , we calculated the alignment coverage , defined as α ( x , y ) = 2la/ ( lx+ly ) , where lx and ly are the length of sequences x and y , and la is the length of the optimal local alignment , define to be the number of columns needed to represent it; that is , it includes gapped positions . The length of the optimal alignment between query x and match y will not , in general , be the same as the length of the optimal alignment between query y and match x . We forced the alignment coverage to be symmetric by setting both α ( x , y ) and α ( y , x ) to the maximum of the two values . By considering only the optimal alignment , we risk underestimating the extent of similarity between homologous sequences . To take suboptimal alignments into account , we used a simple heuristic method for selecting a set of high-scoring local alignments that do not conflict . Two alignments conflict if they overlap or do not appear in the same order in both sequences ( see Text S1 ) . Classifier performance was evaluated using Receiver Operating Characteristic ( ROC ) , which captures the tradeoff between sensitivity ( Sn ) and specificity ( Sp ) as a function of the classifier threshold . A ROC curve is a plot of Sn as a function of 1−Sp , where Sn = TP/ ( TP+FN ) and Sp = TN/ ( TN+FP ) . TP , FP , TN , and FN refer to the number of True Positives , False Positives , True Negatives , and False Negatives , respectively . In the context of our test , TP is the number of sequence pairs that have common ancestry and have been correctly identified by the classifier . FP represents the number of pairs that are classified as homologs , but are not family pairs . TN and FN refer to the number of non-homologous pairs that are correctly ruled out and incorrectly included , respectively . The area under the ROC curve provides a single measure of classification accuracy , corresponding to the fraction of correctly classified entities given the best possible choice of threshold . We used the ROC-n score , defined to be the area under the ROC curve truncated after the first n false positives or ( 4 ) where ti is the number of FF pairs observed before the ith FO pair and T is the total number of FF pairs in the dataset . When the number of negative examples far exceeds the number of positive examples , as is the case here , the ROC score approaches one , resulting in an unjustifiably optimistic assessment of classifier performance . Rn is a more sensitive figure of merit than the untruncated ROC score in this case [78] . We selected n = 100k , where k is the number of FF pairs . This is equivalent to 100 false positives per query . We found that 100k was sufficiently large so that few FF pairs were missed in most tests but not so large so as to obscure the differences in performance between classifiers . The statistical significance of the difference between the ROC-n scores obtained by Neighborhood Correlation and sequence similarity was estimated using p-values calculated using the method described in Schaffer et al . [62] . This method tests the null hypothesis that the difference in ROC-n scores is due the sampling process used to obtain the test data . Rejection of the null hypothesis indicates that the difference in ROC-n scores represents a true difference in the performance of the classifiers . Precision and Recall are also used for evaluation . In the context of our test , Recall denotes the fraction of homologous pairs retrieved and is equivalent to sensitivity . Precision refers to the fraction of protein pairs retrieved that are actually homologous pairs . http://www . neighborhoodcorrelation . org The accession numbers used in this paper are from Swiss Prot ( http://www . ebi . ac . uk/swissprot ) : human PDGFRG ( P09619 ) , human PRKG1B ( P14619 ) , and mouse NCAM2 ( O35136 ) . Accession numbers for all 1577 sequences in the twenty families in our benchmark are given in Dataset S1 . | New genes evolve through the duplication and modification of existing genes . As a result , genes that share common ancestry tend to have similar structure and function . Computational methods that use common ancestry have been extraordinarily successful in inferring function . The practice of discerning evolutionary relationships is stymied , however , by modular sequences made up of two or more domains . When two genes share some domains but not others , it is difficult to distinguish a case of common ancestry from insertion of the same domain into both genes . We present a formal framework to define how multidomain genes are related , and propose a novel method for rapid , robust characterization of evolutionary relationships . In an empirical comparison with the current state of the art , we demonstrate superior performance of our method using a large hand-curated set of sequences known to share common ancestry . The success of our method derives from its unique ability to infer evolutionary history from local topology in the sequence similarity network . This represents a departure from the view that protein family classification must be restricted to families with conserved architecture . By exploiting the structure of the sequence similarity network , our approach surmounts this limitation and opens the door to studies of the role of modularity in protein evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics",
"computational",
"biology/comparative",
"sequence",
"analysis",
"computational",
"biology/protein",
"homology",
"detection",
"computational",
"biology/evolutionary",
"modeling",
"evolutionary",
"biology/genom... | 2008 | Sequence Similarity Network Reveals Common Ancestry of Multidomain Proteins |
Apoptotic cells in animals are engulfed by phagocytic cells and subsequently degraded inside phagosomes . To study the mechanisms controlling the degradation of apoptotic cells , we developed time-lapse imaging protocols in developing Caenorhabditis elegans embryos and established the temporal order of multiple events during engulfment and phagosome maturation . These include sequential enrichment on phagocytic membranes of phagocytic receptor cell death abnormal 1 ( CED-1 ) , large GTPase dynamin ( DYN-1 ) , phosphatidylinositol 3-phosphate ( PI ( 3 ) P ) , and the small GTPase RAB-7 , as well as the incorporation of endosomes and lysosomes to phagosomes . Two parallel genetic pathways are known to control the engulfment of apoptotic cells in C . elegans . We found that null mutations in each pathway not only delay or block engulfment , but also delay the degradation of engulfed apoptotic cells . One of the pathways , composed of CED-1 , the adaptor protein CED-6 , and DYN-1 , controls the rate of enrichment of PI ( 3 ) P and RAB-7 on phagosomal surfaces and the formation of phagolysosomes . We further identified an essential role of RAB-7 in promoting the recruitment and fusion of lysosomes to phagosomes . We propose that RAB-7 functions as a downstream effector of the CED-1 pathway to mediate phagolysosome formation . Our work suggests that phagocytic receptors , which were thought to act specifically in initiating engulfment , also control phagosome maturation through the sequential activation of multiple effectors such as dynamin , PI ( 3 ) P , and Rab GTPases .
An engulfing cell recognizes an apoptotic cell through phagocytic receptor ( s ) and extends thin pseudopods around it to generate a phagocytic cup . The fusion of growing pseudopods and scission of a vacuole containing the apoptotic cell from the plasmalemma generate a phagosome inside the host cell ( Figure 1A ) [1] . The swift engulfment ( phagocytosis ) and subsequent degradation of apoptotic cells inside phagosomes eliminate dying cells before they release any potentially harmful contents , and actively prevent tissue injury , inflammatory responses , and autoimmune diseases [2] . Inefficient degradation of the components of engulfed apoptotic cells , such as nuclear DNA , similar to inefficient engulfment , results in severe inflammatory and autoimmune responses [3 , 4] . Despite its importance , the molecular mechanisms driving the degradation of apoptotic cells remain largely unknown . Our current knowledge of phagosome maturation , the process that results in the complete degradation of phagosomal contents , is primarily obtained through the studies of phagosomes containing opsonized foreign pathogens or undigestible latex beads . Early during phagosomal maturation , signaling molecules such as Rab5 and PI ( 3 ) P are observed to be enriched on nascent phagosomes and act to recruit multiple downstream factors [5] . Early and late endosomes and lysosomes continue to fuse with phagosomes , and provide necessary lipids and proteins for phagosome maturation . Phagosome lumen is gradually acidified . The final state , known as a phagolysosome , is characterized by its abundance of acid hydrolases and an acidic pH environment [5 , 6] . The maturation of phagosomes containing apoptotic cells , by contrast , is less characterized , and may display distinct mechanistic features . Macrophages that internalize foreign pathogens secrete proinflammatory cytokines that induce inflammatory responses , whereas those engulfing apoptotic cells secrete anti-inflammatory cytokines , indicating that the same type of phagocytes elicit different responses to different phagocytic targets [2 , 7] . Rab GTPases are important regulators of many vesicle trafficking events . Rab7 has been implicated in the endocytic pathway , lysosomes biogenesis , and phagosomal maturation [5] . In both mammalian and Dictyostelium cells that ingest latex beads , the overexpression of a dominant-negative form of Rab7 ( Rab7 ( DN ) ) blocks the phagosomal acquisition of lysosomal contents [8 , 9] . It remains to be elucidated , however , whether the defect in phagolysosome formation is a secondary consequence of the defects in lysosomal biogenesis and/or functions . In addition , although the enrichment of Rab7 was observed on phagosomes containing apoptotic cells [10] , whether Rab7 is required for the processing of engulfed apoptotic cells and how Rab7 is activated by upstream signals remain unknown . During the development of Caenorhabditis elegans hermaphrodites , 131 somatic cells and 300–500 germ cells undergo programmed cell death [11] . These apoptotic cells ( referred to as cell corpses ) are distinguishable from living cells as highly refractile discs under a Nomarski differential interference contrast ( DIC ) microscope [12] . In C . elegans , cell corpses are removed by their neighboring cells [11] . Genetic screens have identified two parallel and partially redundant pathways that regulate the engulfment of cell corpses ( Figure 1B ) [1 , 11 , 13 , 14] . In one pathway , the phagocytic receptor cell death abnormal 1 ( CED-1 ) is an engulfing cell-specific , single-pass transmembrane protein that recognizes neighboring cell corpses , clusters to the growing phagocytic cups , and initiates pseudopod extension [15] . The large GTPase DYN-1 ( dynamin ) acts downstream of CED-1 and CED-6 , a candidate adaptor for CED-1 , to control the delivery of intracellular vesicles to both phagocytic cups and phagosomes [1] . In the other pathway , CED-5/Dock180 , CED-2/CrkII , and CED-12/ELMO1 act together to regulate CED-10/Rac1 GTPase and promote cell surface extension during engulfment [11] . CED-10 may also mediate certain activities for the CED-1 pathway [16] . Previously , transmission electron microscopy ( TEM ) analyses indicated that those cell corpses that persisted for hours in ced-1 , -6 , -7 , -2 , -5 , -10 , and -12 mutants were mostly unengulfed . It was thus proposed that these ced genes specifically control the engulfment of cell corpses [15 , 17 , 18] . However , none of the corresponding ced mutants completely block engulfment , since even in the strongest single engulfment mutants , such as ced-1 or ced-5 mutants , only approximately 25% of embryonic apoptotic cells persist for longer than 5 h [1] . Previous TEM analyses , which only characterized the terminal mutant phenotypes hours after the apoptosis events , were unable to determine whether the rest of cell corpses are engulfed , or to detect defects in the kinetics of engulfment and/or phagosome maturation . Recently , we developed a time-lapse imaging protocol and used it to characterize the function of DYN-1 in developing embryos [1] . We found that in engulfing cells , DYN-1 was recruited to the surface of phagocytic cups and phagosomes in a CED-1– , CED-6– , and CED-7–dependent manner . DYN-1 promoted the recruitment and fusion of endosomes to phagocytic membranes , events that provide lipid and protein material to support pseudopod extension and phagosome maturation [1] . These results suggest that the engulfment and degradation of apoptotic cells may share common regulatory mechanisms . Like dyn-1 mutants , ced-1 null mutant embryos display defects in the incorporation of early endosomes to phagocytic cups and phagosomes [1] , suggesting that CED-1 , and perhaps other pathway components , may also control phagosome maturation in addition to engulfment . To test the above hypothesis and to further identify the mechanism controlling the degradation of cell corpses , we developed multiple markers and further expanded the scope of our time-lapse imaging assays to monitor the processes of engulfment and phagosome maturation as well as the cellular events occurring during these processes in living embryos . We identified novel functions of both engulfment pathways in regulating the degradation of engulfed cell corpses . We further elucidate that the signaling pathway led by CED-1 promotes phagolysosome formation through activating RAB-7 , whose function is essential for promoting the recruitment and fusion of lysosomes to phagosomes .
We examined the maturation process of phagosomes containing apoptotic cells in C . elegans . Previously , we reported the transient recruitment of DYN-1 and the incorporation of early endosomes , which are labeled with HGRS-1 , the worm homolog of mammalian endosomal protein Hrs , to the surface of phagosomes [1] . Here , we examined two additional molecular events on phagosomal surfaces: the accumulation of PI ( 3 ) P [19 , 20] and the recruitment of small GTPase Rab7 [8] . We monitored both events in developing embryos using modified versions of an established protocol ( Materials and Methods ) [1] . In all time-lapse experiments described below , we chose to follow the engulfment and degradation of three particular apoptotic somatic cells , C1 , C2 , and C3 , which are located near each other on the ventral side of an embryo and die almost simultaneously during embryogenesis , at approximately 330 min past the first cell division ( the first cleavage ) ( Figure 1C ( l ) and 1D ( a ) ) . ABplaapppa , ABpraapppa , and ABplaapppp , three ventral hypodermal cells , engulf C1 , C2 , and C3 , respectively ( Figure 1C ( f ) and 1D ( a ) ) , during their extension to the ventral midline [1] . In addition , all green fluorescent protein ( GFP ) or monomeric red fluorescence protein ( mRFP1 ) [21] tagged reporters were expressed specifically in engulfing cells under the control of Pced-1 , the ced-1 promoter [15] . The FYVE domain of C . elegans EEA-1 , in a tandem repeat , specifically associates with PI ( 3 ) P [22] . The FYVE-FYVE::mRFP1 reporter was detected primarily in cytoplasm as bright puncta , consistent with its endosomal localization ( Figure 1C ) [1] . The 2xFYVE markers were also evenly distributed in nuclei ( Figure 1D ( d ) ) . In engulfing cells , we observed bright mRFP1 circles surrounding cell corpses ( Figure 1C and 1D ) . Consistent results have been obtained independently [23] . In embryos that coexpressed Pced-1 ced-1::gfp and Pced-1 2xFYVE::mrfp1 , CED-1::GFP , but not 2xFYVE::mRFP1 , was enriched on extending pseudopods ( Figure 1C ) . Strikingly , bright mRFP1 circles appeared on phagosomal surfaces within 4 . 1 ± 1 . 0 min ( n = 16 ) after the closure of pseudopods ( Figure 1C ( d ) , 1C ( j ) , and 1E ) , indicating that PI ( 3 ) P is specifically presented on the surface of nascent phagosomes . Whereas the CED-1::GFP signal rapidly disappeared from phagosomal surface after the enclosure of the phagocytic cup ( within 8 . 4 ± 2 . 0 min , n = 17 ) ( Figure 1C and 1E ) , PI ( 3 ) P remained on a phagosome until its content was completely degraded ( Figure 1D and 1E ) . PI ( 3 ) P thus labels the surface of phagosomes for almost their entire duration . C . elegans rab-7 encodes RAB-7 , a likely ortholog of human Rab7 ( Figure S1A and Text S1 ) . In wild-type animals expressing Pced-1 gfp::rab-7 , a functional GFP::RAB-7 was detected in the cytoplasm , and a portion of it is enriched on cytoplasmic puncta ( Figures S1B and S3D ) . This punctate localization pattern is consistent with previous reports indicating that RAB-7 is localized to late endosomes and lysosomes [24 , 25] . In both embryos and adult hermaphrodite gonads , we observed robust GFP::RAB-7 signals around cell corpses ( Figure S1C ) . Quantitative measurements of time-lapse images ( Materials and Methods ) indicated that the intensity of the GFP signal on phagosomal surfaces could reach as high as 2 . 5 times that in the cytosol of the same engulfing cell ( Figure 1D ( o ) ) . In a time-lapse recording of embryos that coexpress GFP::RAB-7 and 2xFYVE::mRFP1 , RAB-7 was recruited to the surface of a phagosome approximately 3 min after PI ( 3 ) P appeared on the same phagosome ( Figure 1D and 1E ) . After the completion of engulfment , yet prior to the recruitment of RAB-7 , a nascent phagosome appeared as a dark hole surrounded by evenly distributed GFP::RAB-7 in the engulfing cell cytoplasm ( Figure 1D ( h ) and 1D ( i ) ) . Like 2xFYVE::mRFP1 , GFP::RAB-7 persisted on the surface of a phagosome until the phagosome disappeared ( Figure 1D and 1E , and Video S1 ) . To determine whether the enrichment of RAB-7 and PI ( 3 ) P was restricted to phagosomes containing C1 , C2 , and C3 , we examined wild-type 1 . 5-fold to 2-fold stage embryos , which were at 420–460 min after first cleavage [1] . We found that 91% and 73% of cell corpses distinguished under DIC optics are labeled with GFP::RAB-7 and 2xFYVE::GFP , respectively ( Figures 2C ) . These results indicate that both PI ( 3 ) P and RAB-7 are enriched on the surface of all phagosomes , and suggest that they may play important roles during phagosome maturation . Previously , whether C . elegans RAB-7 is involved in cell-corpse removal was unknown . To determine the function of RAB-7 in the degradation of cell corpses suggested by its specific phagosomal localization , we characterized rab-7 ( ok511 ) , a deletion allele obtained from the C . elegans Gene Knockout Consortium ( http://www . wormbase . org ) . We confirmed the presence of a genomic deletion that eliminates the first three exons and a 265-bp upstream sequence of rab-7 in this allele ( Figure 3A ) . Since no additional in-frame ATG codon exists flanking or downstream of the site of deletion , rab-7 ( ok511 ) is likely a null allele . Homozygous rab-7 ( ok511 ) ( m+z−; where m represents the maternal gene product and z the zygotic gene product ) embryos produced by rab-7 ( ok511 ) /+ mothers develop normally and do not contain excess cell corpses ( Figure 3C ) . However , after reaching the adult stage , their gonads contain a large number of germ-cell corpses ( Figure 3B and 3C ) . In contrast , due to the swift removal of cell corpses , the gonads of wild-type hermaphrodites at the same age contain very few cell corpses despite the continuous occurrence of apoptosis events in the germline ( Figure 3B and 3C ) . The homozygous rab-7 ( m−z− ) progeny of rab-7 ( ok511 ) ( m+z− ) animals display 98% embryonic lethality and arrest at various embryonic stages ( Figure 3D ) , indicating an essential function for maternal RAB-7 during embryonic development . We further observed various developmental defects in rab-7 ( m−z− ) embryos , including defects in the consumption of yolk and failure in hypodermal cell closure ( Figures 3D ( c ) and S2 , Text S1 , and Video S2 ) . During embryogenesis , 113 cells undergo apoptosis [12] . However , at late 4-fold stage , the final stage before hatching , no cell corpses can be found in wild-type embryos due to their prompt removal ( Figure 3C and 3D ( a ) ) . By contrast , rab-7 ( m−z− ) embryos that arrest at 4-fold stage contain many persistent cell corpses ( Figure 3C and 3D ( d ) ) , again suggesting defects in their removal . To verify that the persistent cell-corpse–like objects are indeed apoptotic corpses , the ced-3 ( n717 ) mutation , which blocks most , if not all , programmed cell deaths [26 , 27] was introduced into the rab-7 ( ok511 ) background . In rab-7 ( ok511 ) ; ced-3 ( n717 ) double mutants , no cell-corpse–like objects were observed in the m+z− adult hermaphrodite gonads or m−z− embryos ( Figure 3C ) , confirming that the cell-corpse–like objects accumulating in rab-7 mutant animals are apoptotic cells . Accumulation of cell corpses may be caused by excessive apoptosis or defects in the removal of cell corpses . To distinguish between these two possibilities , we monitored the generation and duration of germ-cell corpses induced by γ-ray irradiation in rab-7 ( m+z− ) adults ( Materials and Methods ) . We chose a 2-h recording period starting at 3 h after irradiation , since in the wild-type germline , the number of cell corpses reaches its peak 4 h after irradiation [28] . Within this period , we observed similar numbers of germ cells undergoing apoptosis in wild-type and rab-7 ( ok511 ) animals ( average ten and nine apoptotic cells , respectively ) . This result suggests that rab-7 ( ok511 ) mutants are no more susceptible to apoptosis stimuli than wild-type animals . On the other hand , we observed that in rab-7 mutants , all germ-cell corpses analyzed persisted for a significantly longer period of time than those in wild-type animals ( Figure 3E ) , indicating that rab-7 mutant gonads are defective in the prompt removal of germ-cell corpses . To further distinguish whether the persistent germ-cell corpses observed in rab-7 mutants are results of defects in engulfment or degradation , we analyzed germ-cell corpses and their surroundings using TEM ( Materials and Methods ) . Apoptotic germ cells cellularize from the germline syncytium and are quickly engulfed by gonadal sheath cells , a single layer of somatic cells that wrap around germline syncytium [27] . We analyzed serial sections that span the entire diameter of each cell corpse to determine whether a cell corpse is totally inside a sheath cell . In rab-7 ( ok511 ) mutants , all germ-cell corpses observed in three different gonad arms were engulfed by gonadal sheath cells ( Figure 3F ) . The accumulation of engulfed germ-cell corpses suggests that the degradation , and not the engulfment , of cell corpses is blocked . In addition , in rab-7 mutants , 90% of excessive germ-cell corpses are labeled with 2xFYVE::GFP from engulfing cells , respectively ( Figure 2B and 2C ) , again indicating that they persist inside phagosomes . Under TEM , in 59% of corpses in rab-7 mutants , the nuclear envelope and plasma membrane were readily distinguishable , and the nucleolus appeared as a distinct patch with high electron density ( Figure 3F ( c ) and 3F ( e ) ) . These well-preserved cellular characteristics suggest a defect in the initiation of cell-corpse degradation . In the rest of cell corpses , the electron density of osmium tetroxide–stained sections was greatly enhanced , and it was impossible to distinguish nucleoli or any intracellular membranous structures ( Figure 3F ( e ) and 3F ( g ) ) , suggesting that a certain degree of phagosome maturation had occurred . In 1 . 5- and 2-fold stage rab-7 ( m−z− ) embryos , 78% of cell corpses are labeled with the PI ( 3 ) P marker ( Figure 2A and 2C ) , indicating that they are inside phagosomes . Thus , somatic cell corpses in rab-7 embryos , like germ-cell corpses , appear to be engulfed efficiently yet remain undegraded inside engulfing cells . We refer to the accumulation of engulfed cell corpses as a cell-corpse degradation defective ( Ded ) phenotype . To examine whether the defects in the degradation of germ-cell corpses are an indirect consequence of defects in endocytosis , lysosome biogenesis and function , and/or other vesicle transport events that might be affected by the loss of rab-7 function , we analyzed a number of relevant aspects in rab-7 ( ok511 ) ( m+z− ) adult hermaphrodites , which display a strong Ded phenotype , but develop normally and appear normal in general morphology ( see above ) . To examine possible defects in exocytosis and endocytosis , we first characterized the delivery of yolk to oocytes . In adults , yolk , a lipoprotein complex , is secreted from intestinal cells and transported to the gonad where it is internalized by oocytes through receptor-mediated endocytosis [29] . By monitoring YP170 ( a yolk component ) ::GFP [29] , we observed that just as in wild type [1] , 100% of fertilized embryos inside rab-7 ( ok511 ) ( m+z− ) adult hermaphrodite gonads contain internalized YP170::GFP ( Figure 4A ) . This result indicates that in these adults , the exocytosis and the subsequent endocytosis of yolk into oocytes are both normal . The accumulation of germ-cell corpses thus is unlikely to be a secondary consequence of a general endocytosis and/or exocytosis defect . To detect lysosomes in rab-7 mutants , we developed a specific fluorescent lysosomal marker , CTNS-1 , tagged with GFP or mRFP1 at its C-terminus . CTNS-1 is the C . elegans homolog of human lysosomal cystine transporter cystinosin [30 , 31] . CTNS-1 colocalized to cytoplasmic puncta with RAB-7 and LMP-1 , two proteins reported to localize to lysosomes , indicating that CTNS-1 was primarily localized to lysosomal compartments ( Figure S3 and Text S1 ) . We compared the number and morphology of CTNS-1::GFP ( + ) particles in several tissues in wild-type and rab-7 ( m+z− ) adult hermaphrodites and did not observe any obvious difference between these strains ( Figure 4B ) . Recently , it was reported that in rab-7 ( ok511 ) ( m+z− ) adult hermaphrodites , the acidification of lysosomal compartments in coelomocytes , special scavenger cells that are highly active in endocytosis , was normal [32] . These lines of evidence suggest that the biogenesis and functions of lysosomes in rab-7 ( ok511 ) ( m+z− ) adult animals are relatively normal , probably owing to the residual wild-type maternal gene activity . By contrast , in rab-7 ( ok511 ) ( m−z− ) embryos that lack both maternal and zygotic gene activity , we detected severe defects in lysosomal functions ( Figure S2 and Text S1 ) . We thus conclude that the accumulation of engulfed cell corpses observed in the gonads of rab-7 ( ok511 ) ( m+z− ) adult animals is unlikely a secondary consequence of general defects in lysosome biogenesis or function . We observed that in rab-7 ( ok511 ) ( m+z− ) hermaphrodites , coelomocyte vacuoles , which represent endocytic compartments [33] , appeared the same size as in wild type , although they accumulated in a larger number ( Figure 4C ) . This is in sharp contrast to the phenotypes observed in typical lysosome biogenesis mutants , such as cup-5 or ppk-3 mutants , on which coelomocyte vacuoles are enlarged [25 , 33] . On the other hand , mutations in cup-5 or ppk-3 resulted in no or very weak Ded phenotypes ( Figure 4D ) . In addition , in cad-1 mutants , in which the activity of aspartyl protease cathepsin D , a lysosomal hydrolase , was severely reduced [34] , no Ded phenotype was observed ( Figure 4D ) . These results imply that in addition to its function in regulating lysosome biogenesis , RAB-7 mediates an independent function related to phagosome maturation . GFP::RAB-7 , when produced in engulfing cells ( Pced-1gfp::rab-7 ) , efficiently rescued the Ded phenotype in rab-7 ( ok511 ) mutant embryos , L1 larvae , and adult gonad ( Figure 5A ) . On the other hand , a Pegl-1gfp::rab-7 construct , which specifically expressed GFP::RAB-7 in apoptotic cells ( Figure 5B ) [35] , did not display any rescuing activity ( Figure 5A ) . These observations suggest that rab-7 is likely to act in engulfing cells , not dying cells , to promote cell-corpse degradation . Rab proteins are known to cycle between the GTP-bound , active state and the GDP-bound , inactive state [36] . A T22N mutation of mammalian Rab7 ( Figure S1A ) results in a great decrease of its in vitro GTP-binding activity [37]; on the other hand , the Q67L mutation ( Figure S1A ) decreases its GTP hydrolysis activity , and thus locks it in a GTP-bound and active state [38 , 39] . We analyzed these two mutations for their effects on RAB-7 function . The transgene Pced-1gfp::rab-7 ( T23N ) did not rescue the Ded phenotype in rab-7 ( ok511 ) mutants ( Figure 5A ) , indicating that GTP binding is essential for RAB-7 function . On the other hand , Pced-1gfp::rab-7 ( Q68L ) displayed a rescuing activity similar to ( in gonads and embryos ) or stronger than ( in L1 heads ) that of Pced-1gfp::rab-7 ( Figure 5A ) , suggesting that like many other small GTPases , GTP RAB-7 is an active form in cell-corpse degradation . In engulfing cells , GFP::RAB-7 ( Q68L ) , like GFP::RAB-7 , was enriched on the surface of phagosomes ( unpublished data ) . In contrast , GFP::RAB-7 ( T23N ) failed to accumulate to phagosomes ( Figure 5C ) . These results indicate that only the GTP-bound , active form of RAB-7 is recruited to phagosomal surfaces and establish a correlation between the localization and function of RAB-7 on phagosomal surfaces . We observed that in both embryos and gonads , the percentage of cell corpses labeled by 2xFYVE::GFP circles were similar between wild type and rab-7 mutant ( Figure 2 ) , indicating that the enrichment of PI ( 3 ) P on phagosomal surfaces is normal in rab-7 mutant . In addition , in embryos that in engulfing cells expressed HGRS-1::GFP , a marker for early endosomes [1] , both the percentage of phagosomes labeled with HGRS-1::GFP ( 39% and 42% , respectively ) and the intensity of GFP on phagosomal surfaces were similar between wild type and rab-7 ( ok511 ) ( Figure 2A and 2C ) [1] . This result indicates that the function of RAB-7 is not essential for the delivery of early endosomes to phagosomes . In wild-type embryos , 64% of cell corpses are labeled with CTNS-1::GFP on their surfaces ( Figure 6A ) , indicating that lysosomes are incorporated into phagosomes . We monitored the delivery of CTNS-1::GFP signal to phagosomal surfaces in both embryos that expressed Pced-1 ctns-1::gfp and embryos that coexpressed Pced-1 gfp::rab-7 and Pced-1 ctns-1::mrfp1 . The dynamic CTNS-1::GFP recruitment to phagosomes occurred in two distinct phases . During phase I , which started after the formation of nascent phagosomes and lasted approximately 20 min , a weak and punctate CTNS-1::GFP signal was observed on the surface of phagosomes ( Figure 6B ( g–i ) , 6C ( a–d ) , and 6C ( o ) ) . During the following phase II , CTNS-1::GFP rapidly accumulated on a phagosome , and gradually evolved from multiple punctate spots to a smooth , continuous circle ( Figure 6B ( i–l ) and 6C ( d–g ) ) . Following the rapid enrichment of CTNS::GFP signal , the volume of a phagosome quickly decreased ( Figure 6B , 6C ( o ) , and 6C ( p ) ) , indicating degradation of phagosomal contents . A phagosome remained CTNS-1::GFP ( + ) until it was completely degraded ( Figure 6B and 6C ) . The accumulation of the CTNS-1::GFP signal is indicative of the recruitment of lysosomes to phagosomal surfaces , whereas the subsequent transition from punctate to continuous circles may represent the fusion of lysosomes to phagosomal membranes . The recruitment of RAB-7 to phagosomes took place earlier than the phase II CTNS-1 enrichment ( Figures 1E and 6B ) , with an average interval of 13 . 4 ± 5 . 8 min ( n = 9 ) . In rab-7 ( ok511 ) embryos , as in wild type , CTNS-1::GFP is localized to cytoplasmic puncta ( Figures 6A ) . However , we observed severe defects in the incorporation of lysosomes into phagosomes . Most ( 65% ) of all CTNS-1::GFP circles observed around cell corpses in 1 . 5-fold rab-7 ( ok511 ) embryos remain punctate ( Figure 6A ( e ) ) . Time-lapse recording results indicated that the localization of CTNS-1::GFP to phagosomes was mostly arrested at phase I: the signal intensity on phagosomes remained low ( Figure 6C ( o ) ) , suggesting a defect in the recruitment of lysosomal particles; furthermore , the signal observed on phagosomal surfaces remained in a punctate pattern for an extended period of time ( >100 min ) ( Figure 6C ( h–n ) ) , suggesting a likely defect in lysosomes/phagosome fusion . To quantify the degradation defect caused by the rab-7 ( ok511 ) mutation , we monitored the duration of CTNS-1::GFP-labeled phagosomes . The volume of phagosome C1 in wild-type embryo decreased 5-fold within 40 min of its presence , yet remained unchanged over 2 h in a rab-7 ( ok511 ) embryo ( Figure 6C ( p ) ) , indicating a lack of digestion of phagosomal contents . In wild-type embryos , phagosomes disappeared within a period of 30–70 min after their formation; in rab-7 ( ok511 ) embryos , ten phagosomes monitored all lasted much longer than 90 min ( Figure 6D ) . These results indicate that the defects in the incorporation of lysosomes to phagosomes caused by rab-7 deletion are closely associated with the blockage of the degradation of phagosomal contents . We further monitored individual CTNS-1::GFP ( + ) puncta over time and observed that in wild-type embryos , the CTNS-1 ( + ) puncta were delivered to the surfaces of phagosomes through either direct encounters with phagosomes or attachment to membrane tubules originating from phagosomes . Tubules extending from phagosomal surfaces were frequently observed; these tubules were labeled with RAB-7 and CTNS-1 ( Figures 6B and 7A ) , but not with PI ( 3 ) P ( Figure 1D ) . As shown in Figure 7A and Video S3 , within 6 min , a CTNS-1 ( + ) punctum was seen to attach to such a tubule , and was carried to the phagosome by the retracting tubule . In rab-7 ( ok511 ) embryos , in contrast , we did not observe any tubular structures extending from phagosomes ( Figure 7 and Video S4 ) . The lack of extended tubules presumably leads to the decreased targeting of lysosomes to phagosomes , which subsequently results in insufficient delivery of lysosomal enzymes . Indeed , on the surface of 43% of phagosomes that we monitored in rab-7 ( ok511 ) embryos , no CTNS-1 ( + ) particles were ever observed to be attached during the entire 100-min observation period ( Figure 7B and Video S4 ) . These observations suggest that RAB-7 is essential for the extension of membrane tubular structures from phagosomes , a process necessary for the efficient recruitment of lysosomes to the surface of phagosomes . In wild-type embryos , after attaching to the phagosomal surface , lysosomal particles rapidly lost their distinct shape and became part of the phagosomal surface , a process likely suggesting the fusion of lysosomal particles to phagosomal membranes ( Figures 6C ( a–g ) and 7A ) . In rab-7 ( ok511 ) embryos , however , in some cases , even after the attachment to phagosomal surfaces , the particles maintained their distinct shape for various amounts of time , sometimes greater than 50 min ( Figure 7A ( i–p ) and Video S4 ) , indicating an apparent lack of fusion . The defects in the attachment and possible fusion of individual puncta to phagosomes apparently correlate with the persistent punctate state and the low signal intensity of CTNS-1::GFP on phagosomal surfaces ( Figure 6C ( h–n ) ) . Previously , due to the lack of markers to label phagosomes and time-lapse imaging techniques to follow phagosome maturation , the ced genes acting in the two parallel pathways led by ced-1 and ced-5 were only implicated in the engulfment of cell corpses ( Introduction ) . The diffused cytoplasmic localization pattern of GFP::RAB-7 , in addition to its association to cytoplasmic puncta , allows us to follow the extension and closure of pseudopods during engulfment ( Figure S4A and S4B ) . In addition , a phagosome can be distinguished by the enriched GFP::RAB-7 signal on its surface ( Figure 8A ( b ) ) , or even before being labeled by GFP::RAB-7 , recognized as a dark , GFP ( − ) hole inside a GFP::RAB-7 ( + ) engulfing cell ( Figure 8A ( a ) ) . To further study whether the engulfment and degradation of cell corpses employ common mechanisms , we examined the kinetics of the engulfment and degradation in ced-1 , dyn-1 , and ced-5 mutant embryos using time-lapse imaging . ced-1 and ced-5 mutant embryos are viable [13] . Although dyn-1 ( n4039 ) embryos undergo developmental arrest , the arrest point is at 4-fold stage ( >620 min after first cleavage ) [1] . Thus our time-lapse recording , performed between 300–420 min after first cleavage , was not affected by the arrest . We monitored the engulfment of cell corpses C3 in ced-1 ( e1735 ) , dyn-1 ( n4039 ) , and ced-5 ( n1812 ) null mutant embryos [1 , 15 , 40] that express Pced-1 gfp::rab-7 . We demonstrated that these three mutants displayed distinct and severe defects in the initiation and/or the extension of pseudopods around cell corpses ( Figure S4B–S4E and Text S1 ) , consistent with the well-characterized functions of these proteins in initiation of engulfment . Interestingly , in only 19%–25% of single-mutant embryos , C3 remained unengulfed for longer than 100 min; in the other embryos , C3 were eventually engulfed ( Figure S4E ) . In addition , in 1 . 5-fold stage ced-1 and ced-5 mutant embryos , approximately 20%–30% of DIC ( + ) cell corpses appeared as either dark holes or GFP-labeled phagosomes inside engulfing cells ( Figure S4F and S4G ) , indicating that these cell corpses were engulfed , but not yet degraded . The partial defects in engulfment allowed us to monitor the degradation of those cell corpses that were engulfed in ced-1 , dyn-1 , and ced-5 embryos . In wild-type embryos , the duration of a phagosome , which was the total time period that it existed as a dark hole ( nascent phagosome ) and then as a GFP::RAB-7 ( + ) circle , was 30–70 min ( Figure 8A ( a–e ) and 8B ) . In ced-1 , dyn-1 , and ced-5 mutants , however , the duration varied within a wider range , with a distinct population of phagosomes that lasted longer than 90–100 min , the time limit of our time-lapse recording due to the vigorous body movement of embryos occurring after reaching approximately 440 min after first cleavage ( Figure 8B ) . The volume of the phagosomes belonging to this population did not reduce throughout the observation period ( Figure 8A ( v ) ) , indicating a severe defect in the degradation of phagosomal content . This defect is further confirmed using CED-1C ( the intracellular domain of CED-1 ) ::GFP , another reporter that is evenly distributed in the cytoplasm of engulfing cells and allows the recognition of phagosomes as GFP ( − ) holes ( Figures 8C and S5 ) . The above observations confirm the defects of dyn-1 mutants in phagosome maturation , which was observed before using different assays [1] , and further reveal novel functions for ced-1 and ced-5 in the degradation of apoptotic cells , a step downstream of engulfment . Since the formation of phagolysosomes is an essential step in the digestion of phagosomal contents ( Figure 6 ) , we first analyzed the incorporation of lysosomes to maturing phagosomes containing C1 , C2 , and C3 in ced-1 , dyn-1 , and ced-5 embryos by monitoring CTNS-1::GFP . We observed that although the initial localization of weak CTNS-1::GFP ( + ) signal on phagosomal surfaces in phase I was relatively normal; the enrichment of CTNS-1::GFP signal expected for phase II was delayed or absent on the majority of phagosomes in all three mutants ( Figure 9 ) . In many cases , the weak CTNS-1::GFP signal on phagosomal surfaces was unstable ( Figure 9A ) . The defects of lysosomes incorporation in ced-1 and dyn-1 mutant embryos are more severe than that in ced-5 embryos ( Figure 9D ( a ) ) , correlating with the levels of degradation defects displayed by the corresponding mutants ( Figure 8B ) . Although the recruitment of CTNS-1 ( + ) particles was severely delayed , once recruitment was complete , in most cases , the corresponding phagosome would disappear within 50 min ( Figure 9B and 9D ( b ) ) . For instance , we observed different fates of two phagosomes in one dyn-1 ( n4039 ) embryo: the weak , initial CTNS-1 signal labeling the phagosome containing C3 failed to increase over time , and C3 remained undegraded after 65 min ( Figure 9A ) , whereas the phagosome containing C1 was eventually degraded after a delayed enrichment of CTNS-1 signal ( at 31 min ) on its surfaces ( Figure 9B ) . These results suggest that the inefficient incorporation of lysosomes into phagosomes is one of the major causes for the cell-corpse degradation defects observed in ced-1 , dyn-1 , and ced-5 mutants . Since RAB-7 activity is essential for the incorporation of lysosomes to phagosomes , we monitored the dynamic association of RAB-7 with phagosomes in ced-1 , dyn-1 , and ced-5 mutants . In both wild-type and ced-5 embryos , greater than 70% of phagosomes are labeled with a GFP::RAB-7 circle within 10 min after the completion of engulfment ( Figure 8D ) . By contrast , in ced-1 and dyn-1 embryos , the recruitment of RAB-7 to phagosomal surfaces is delayed or blocked . The majority of phagosomes in ced-1 ( 82% ) and dyn-1 ( 79% ) mutants remained RAB-7 ( − ) ( detected as dark holes ) for greater than 10 min ( Figure 8A ( f–t ) and 8D , and Videos S5–S7 ) . In particular , 35% and 17% of phagosomes in ced-1 and dyn-1 mutant embryos , respectively , remained RAB-7 ( − ) for greater than 60 min ( Figure 8A ( k–t ) and 8D ) ; all 11 such phagosomes failed to reduce in size throughout the recording period ( for instance , Figure 8A ( p–t ) ) . These phenotypes indicate that ( 1 ) the recruitment of RAB-7 to phagosomal surfaces is essential for phagosome maturation , and ( 2 ) the recruitment is inefficient in ced-1 and dyn-1 mutants but relatively normal in ced-5 mutants . We examined the localization of RAB-7 ( Q68L ) in ced-1 ( e1735 ) mutant embryos and found that the recruitment of RAB-7 ( Q68L ) to phagosome surface is also delayed . However , the defect is less severe: whereas greater than 30% of phagosomes took more than 60 min to recruit RAB-7 after engulfment completes , all phagosomes recruit RAB-7 ( Q68L ) within 50 min , with the majority within 30 min ( Figure S6 ) . These results suggest that enhancing GTP binding appears to alleviate the defect of RAB-7 recruitment caused by the loss of ced-1 function . To identify additional events important for phagosomal maturation regulated by the CED-1 pathway , we monitored the presentation of PI ( 3 ) P on phagosomes , which occurs prior to the recruitment of RAB-7 in wild-type ( Figure 1D ) , ced-1 ( e1735 ) , ced-5 ( n1812 ) , ced-6 ( n2095 ) ( a strong loss-of-function allele ) [41] , and dyn-1 ( n4039 ) embryos using the 2xFYVE::GFP or ::mRFP1 markers . In time-lapse experiments , nascent phagosomes not yet labeled with 2xFYVE markers were distinguished as dark holes inside engulfing cells ( Figure 10A ( b ) ) . We observed the delays of PI ( 3 ) P presentation to different extents in different mutant embryos . In wild-type embryos , all phagosomes were labeled with 2xFYVE::GFP within 10 min after the completion of engulfment , among which 87% were labeled within 6 min ( Figures 1C and 10C ) . The ced-5 ( n1812 ) mutation resulted in a slight delay of the appearance of PI ( 3 ) P: on 62% of phagosomes , PI ( 3 ) P was detected within 11–20 min after engulfment ( Figures 10C and S7C , and Video S8 ) . ced-6 ( n2095 ) embryos displayed an intermediate delay of PI ( 3 ) P presentation ( Figures 10C and S7A ) . The ced-1 and dyn-1 embryos displayed the most severe defects—only 14% and 8% of phagosomes , respectively , were labeled with PI ( 3 ) P within 10 min of engulfment ( Figure 10A–10C , and Videos S9 and S10 ) . In addition , PI ( 3 ) P was absent from 21% and 42% of phagosomes , respectively , during a period that lasted greater than 60 min , suggesting that the presentation of PI ( 3 ) P on the surface of these phagosomes was blocked ( Figures 10C and S7B , and Video S10 ) . These observations suggest that the severe delay and/or blockage of PI ( 3 ) P presentation in ced-1 and dyn-1 mutants might affect multiple downstream maturation events .
Previously , due to the lack of cellular markers , whether a cell corpse was engulfed could not be easily distinguishable in C . elegans embryos . We have developed a series of live-cell imaging–based assays and have provided the first real-time observation of the entire cell-corpse removal process in the context of a developing embryo . Using these assays , we also established the temporal sequence of multiple phagosome maturation events . These in vivo assays have revealed that C . elegans RAB-7 plays essential roles in the formation of phagolysosomes . Our study indicates that the degradation of apoptotic cells in C . elegans shares certain mechanistic features with that of opsonized foreign objects in mammalian macrophages . Unlike TEM , which only detects the terminal phenotypes , the time-lapse recording system allows us to observe previously unknown defects in the dynamic process of phagosome maturation . We thus identified distinct roles of four C . elegans proteins , CED-1 , CED-6 , DYN-1 , and CED-5 , in the degradation of engulfed apoptotic cells ( Figure 11 ) . CED-1 , CED-6 , and CED-5 were only known for their functions in engulfment ( see Introduction ) . Whether the mammalian homologs of these proteins are similarly involved in the maturation of phagosomes remains to be examined . Our findings suggest that C . elegans is a powerful model system for identifying additional genes and new mechanisms regulating phagosome maturation . Mammalian macrophages elicit pro- or anti-inflammatory responses after engulfing pathogens , which are considered “foreign , ” or apoptotic cells , which are considered “self , ” respectively ( see Introduction ) . The mechanisms behind these distinct responses remain largely unknown , and phagosomal surfaces are possible sites where these immune responses could be initiated . By studying the removal of apoptotic cells in C . elegans , we may find clues to those unknown mechanisms . For example , we observed that PI ( 3 ) P was enriched on C . elegans nascent phagosomes and persisted there until the complete digestion of phagosomal content . On the contrary , PI ( 3 ) P was reported to be transiently present on the surface of mammalian phagosomes containing opsonized objects , disappearing well before the completion of phagosome maturation [20] . These observations may reflect differences in the maturation of phagosomes with different contents . C . elegans rab-7 was implicated in controlling cell survival based on the observation that rab-7 ( RNA interference [RNAi] ) resulted in excessive germ-cell corpses in the adult hermaphrodite gonads [42] . Here , we report multiple lines of evidence demonstrating that the excessive germ-cell corpses observed in rab-7 null mutants are unlikely caused by excessive cell death; rather , they are the result of defects in the degradation of engulfed apoptotic cells ( Figure 3 ) . Mammalian Rab7 is involved in endocytic pathway and lysosome biogenesis [37 , 43–45] . Consistently , C . elegans RAB-7 is essential for endosomal trafficking and yolk consumption ( Figure 11C ) [29 , 32] . Our research provides several lines of evidence to suggest that the Ded phenotype of rab-7 mutants is not merely a secondary consequence of the defects in the aforementioned functions , rather , RAB-7 plays a direct and independent role in phagosome maturation by acting on the surface of phagosomes . In adult rab-7 ( m+z− ) mutant hermaphrodites , whereas intracellular trafficking and lysosome biogenesis are relatively normal due to the activity of rab-7 maternal product , the degradation of engulfed cell corpses is severely impaired . In addition , mutations in ced-1 and dyn-1 cause strong and correlated defects in the recruitment of RAB-7 to phagosomal surfaces and in the formation of phagolysosomes , further supporting an independent role of RAB-7 on phagosomal surfaces . Studies of the mammalian phagosomes containing infectious bacteria indicate that the functions and regulation of host-cell Rab7 during phagosome maturation are complex and diverse , and depend on the ingested bacteria species [46–48] . We found that regarding the degradation of apoptotic cells , which are distinct from latex beads , which are not degradable , and from infectious bacteria , which are foreign objects , RAB-7 displays both conserved and unique features . Mammalian Rab7 and its effector RILP were implicated in the extension of tubular structures from phagosomal surfaces , which attract late endosomes and lysosomes to phagosomes containing latex beads , based on the effects of the overexpressed dominant-negative forms of Rab7 and RILP [8] . We observed a similar extension of tubular structures from phagosomes and the recruitment of CTNS-1 ( + ) lysosomal particles along those tubules in C . elegans ( Figure 7 ) . Furthermore , by analyzing rab-7 null mutant embryos , we discovered that one of the functions of endogenous RAB-7 is to promote the extension of these tubules and the recruitment of lysosomes to phagosomes . We have also identified another function of RAB-7 in promoting the fusion of lysosomes to phagosomal surfaces . This fusion activity is consistent with the reported activity of Ypt7 , the yeast counterpart of Rab7 , in promoting the physical contact of vacuoles during homotypic vacuole fusion ( reviewed in [49] ) , and indicates that RAB-7 is capable of mediating the fusion of organelles of different origins . In addition , dominant-negative Rab7 inhibits acidification of mammalian phagosomes that contain latex beads [8] . In contrast , we found that the null mutation of rab-7 did not significantly affect the acidification of phagosomes containing apoptotic cells ( Figure S8 and Text S1 ) . Phagosomes containing apoptotic cells thus provide a new experimental system for further studying the mechanisms behind the functions and regulation of Rab7 . In engulfing cells , RAB-7 exists in the cytoplasm both in a diffused form and in association with intracellular organelles , in particular lysosomes . The initial phagosomal appearance of RAB-7 occurs prior to the phagosomal acquirement of lysosomes , suggesting that free , cytosolic RAB-7 molecules are likely the primary source for phagosome-associated RAB-7 . The GTP-bound form of RAB-7 is the active form recruited to and stably associated with phagosome , perhaps through a particular protein or protein complex that that mediates this association . We have identified the novel functions of at least two genes , ced-1 and dyn-1 , for the timely recruitment of RAB-7 to phagosomes . This finding , together with the finding that RAB-7 is essential for the recruitment and fusion of lysosomes into phagosomes , indicate that the CED-1 pathway controls phagolysosome formation through regulating RAB-7 recruitment and function . Previously , we have reported that the phagocytic receptor CED-1 acts upstream of DYN-1 in the signaling pathway for engulfment [1] . We have also established that the transient localization of DYN-1 on the surface of phagocytic cups and phagosomes are both dependent on CED-1 function [1] . Thus we propose that CED-1 acts upstream of DYN-1 to control RAB-7 recruitment ( Figure 11B ) . The molecular mechanism for the recruitment of RAB-7 remains to be investigated . The observation that the recruitment of a GTP-bound , constitutively active form of RAB-7 to phagosomal surface was less affected than that of RAB-7 by the ced-1 null mutation ( Figure S6 ) may indicate that the CED-1 pathway may induce the formation of GTP-RAB-7 around phagosomal surfaces and the subsequent RAB-7/phagosome association . Alternatively , the CED-1 pathway may act in a more indirect manner , by recruiting or activating early signaling molecules that will further recruit RAB-7 ( Figure 11A ) . We observed that PI ( 3 ) P was dramatically enriched on phagosomal surfaces prior to RAB-7 . Furthermore , in dyn-1 , ced-6 , and ced-1 mutants , the enrichment of PI ( 3 ) P on phagosomal surfaces was delayed . These observations suggest that PI ( 3 ) P might be related to the recruitment of RAB-7 . The enrichment of RAB-7 on phagosomes is not the only phagosome maturation event regulated by the CED-1 signaling pathway . For instance , the recruitment of early endosomes to phagosomes , which is independent of RAB-7 ( Figure 2 ) , is dependent on CED-1 , CED-6 , and DYN-1 [1] . In addition , we observed that the rate of enrichment of PI ( 3 ) P on phagosomal surfaces is under the control of ced-1 , ced-6 , and dyn-1 . PI ( 3 ) P attracts FYVE or PX domain–containing effectors , some of which play diverse roles during phagosome maturation , to phagosomal surfaces ( reviewed in [50] ) . CED-1::GFP was observed to transiently cluster on nascent phagosomal surfaces for merely 9 min , in a time period that barely overlaps with that of PI ( 3 ) P and RAB-7 ( Figure 1E ) . However , DYN-1 remains on phagosomal surfaces for a longer period ( 18 min ) . CED-1 thus may establish an active state of phagosomal surfaces through the activation of DYN-1 . The rapid accumulation of PI ( 3 ) P on the surfaces of nascent phagosomes is likely to reflect the local activation of a Class III PI ( 3 ) kinase [20] , whose activity and/or localization on phagosomal surfaces may be controlled by the CED-1 pathway . We thus propose that CED-1 , CED-6 , and DYN-1 act in a signaling pathway to control both the engulfment and the degradation of engulfed apoptotic cells , through regulating different downstream effectors ( Figure 11B ) . On nascent phagosomes , or perhaps even before the phagocytic cup is enclosed , this pathway may actively prime the phagosomal membrane for the reception of multiple signaling molecules and sequential incorporation of intracellular vesicles ( Figure 11A ) . Although extensive studies suggest that nascent phagosomes possess multiple signaling molecules on their surfaces , the identity of the ultimate initial signal that triggers phagosome maturation remains unclear . An activated phagocytic receptor may well be the source of this signal . When in complex with extracellular ligands , phagocytic receptors were known to undergo certain conformational changes that lead to the activation of their intracellular domains [51] , which in turn may act on the cytoplasmic side of phagocytic cups and phagosomes to recruit cytosolic factors . Our report that CED-1 initiates phagosomal maturation through activating downstream effectors suggests that phagocytic receptors for apoptotic cells in different organisms , including mEGF10 , the human CED-1 homolog [52] , and perhaps even phagocytic receptors for foreign pathogens , may act to regulate the downstream phagosome maturation events . Furthermore , by activating different phagocytic receptors , phagocytic targets that are “self” or “foreign” might thus initiate differential phagosome maturation processes . ced-5 null mutants are defective in the recruitment of lysosomes to phagosomes . However , unlike ced-1 or dyn-1 mutations , null mutation in ced-5 does not seem to affect the recruitment of RAB-7 on phagosomal surfaces , nor does it cause any severe defect in the presentation of PI ( 3 ) P on phagosomal surfaces . These results imply that CED-5 acts in a distinct pathway to control phagolysosome formation ( Figure 11B ) , just like in engulfment ( Figure 1B ) . During engulfment , the pathway led by CED-5 was shown to regulate cytoskeletal reorganization ( reviewed in [11] ) . Recently , it was reported that cytoskeletal reorganization also plays an active role in phagosome maturation in mammalian cells [10 , 53] . It would be interesting to know whether CED-5 and other members of its pathway contribute to phagosome maturation through remodeling the cytoskeleton .
C . elegans strains were grown at 20 °C as previously described [54] . The N2 Bristol strain was used as the reference wild-type strain . Mutations and integrated transgenes used are described by [55] , except where noted otherwise: LGI , ced-1 ( e1735 ) ; LGII , rab-7 ( ok511 ) , cad-1 ( j1 ) [34] , mIn1[dpy-10 ( e128 ) mIs14[myo-2::gfp + pes-10::gfp]] [56]; LGIII , cup-5 ( ar465 ) [57]; cup-5 ( n3194 ) [58] , ced-6 ( n2095 ) ; LGIV , ced-5 ( n1812 ) , ced-3 ( n717 ) ; LGV , unc-76 ( e911 ) ; LGX , dyn-1 ( n4039 ) [1] , ppk-3 ( n2668 ) [25] . rab-7 ( ok511 ) was generated and provided by the C . elegans Gene Knockout Consortium in strain VC308 ( genotype: rab-7 ( ok511 ) /mIn1 ) . Transgenic lines were generated by microinjection [59] . Plasmids were coinjected with p76-18B [60] into unc-76 ( e911 ) mutants , and non-Unc progeny were identified as transgenic animals . rab-7 , ctns-1 ( a ) , and lmp-1 cDNAs were PCR amplified from a mixed stage C . elegans cDNA library ( Z . Zhou and H . R . Horvitz , unpublished data ) and confirmed to bear anticipated sequences . The T23N and Q68L mutations were introduced into rab-7 cDNA using a QuikChange Site-directed Mutagenesis Kit ( Stratagene ) . To generate Pced-1gfp::rab-7 , the gfp coding sequence lacking the stop codon , obtained from plasmid pPD95 . 75 ( a gift from A . Fire ) , was fused in frame to the ATG start codon of the rab-7 coding sequence . The fusion gene was cloned into a pPD95 . 75-based plasmid , flanked by Pced-1 , the 5 . 2-kb ced-1 promoter [15] , and the unc-54 3′ UTR on the 5′ and 3′ ends . The mrfp1 cDNA is a gift from R . Tsien [21] . The Pced-1lmp-1::mrfp1 , Pced-1ctns-1 ( a ) ::gfp , or Pced-1ctns-1 ( a ) ::mrfp1 constructs were similarly generated by cloning the lmp-1::mrfp1 , ctns-1 ( a ) ::gfp , or ctns-1 ( a ) ::mrfp1 genes into the above plasmid , flanked by Pced-1 and unc-54 3′UTR on the 5′ and 3′ ends , respectively . The gfp::rab-7 fusion gene was also inserted between Pegl-1 and egl-1 3′ UTR to generate Pegl-1gfp::rab-7 . rab-7 genomic DNA was PCR amplified using VC308 lysate as a template . The fragment amplified from the ok511 allele , which was 742 bp shorter than wild type , was sequenced , and the exact boundaries of the deletion , tgtgatttttc . . . aattcctccaa , as reported by C . elegans Gene Knockout Consortium , was confirmed . Adult hermaphrodites 48 h past the mid-L4 stage were used for TEM analysis of germ-cell corpses and the neighboring gonadal sheath cells as previously described [1] . Serial 50-nm thin sections that cover the entire length of each cell corpse were analyzed to determine whether a germ-cell corpse is entirely inside the sheath cell . Both somatic cell corpses in embryos and germ-cell corpses in the adult hermaphrodite gonads were scored under the Nomarski DIC microscope by their highly refractile appearance as previously established [1] . Embryonic stages were determined according to [1] . In brief , 1 . 5- and 2-fold stages are embryos whose body lengths are 1 . 5 or two times the length of an egg , respectively . Late 4-fold stages embryos are embryos whose bodies make three turns and contain fully developed pharyngeal grinders . Wild-type embryos at 1 . 5- , 2- , and early and late 4-fold stages correspond to embryos at approximately 420 min , 445 min , 660 min , and approximately 780 min past the first cleavage , respectively [1] . Time-lapse analysis was performed on a DeltaVision system with Olympus IX70 microscope ( Applied Precision ) . Adult hermaphrodites 24 h post-L4 stage were irradiated with a 137Cs source ( Gammacell 1000 , 8 . 33Gy/min ) at 180Gy . 2 . 5 h after irradiation , samples were transferred onto NGM plates containing 1 mM Aldicarb; after 20 min , samples were mounted on a slide immersed in 0 . 8 mM Aldicarb solution and observed under DIC optics at 20 °C . Twenty serial Z-sections , at 1 . 0 μm/section , of the gonadal region were recorded every 3 min . Animals were closely monitored for viability during recording . The total number and the period of duration of cell corpses that occur during the first 2 h were scored . An Olympus IX70-Applied Precision DeltaVision microscope equipped with a Photometris Coolsnap digital camera and Applied Precision Softworx software were used to capture and deconvolve fluorescence images . To score the number of phagosomes labeled with a certain fluorescence marker , serial Z-section images of an entire embryo were recorded at 0 . 5-μm intervals , in 40 sections , and images were deconvolved before scoring . To monitor the process of engulfment and degradation of cell corpses C1 , C2 , and C3 , as well as the subcellular localization of multiple cellular markers , recording of the ventral surface of embryos started at 310–320 min after first cleavage and lasted for 2 h in 1- or 2-min intervals . For each time point , eight serial Z-sections at a 0 . 5-μm interval were recorded . Signs such as embryo elongation and movement were closely monitored to ensure the embryo being recorded developed normally . Fluorescence signal intensity was measured and images were analyzed using the ImageJ software . To calculate the ratio of signal intensity on the surface of phagosomes compared to that in the engulfing cell cytosol , the phagosomal surface and cytoplasm were individually defined using the “Free Hand” tool , and the modal value of each selected area was used in the ratio calculation . The enrichment of fluorescence signal on the phagosomal surface was considered significant when the ratio was greater than 1 . 2 . To measure the volume ( V ) of phagosomes , we assumed they were spheres . The radius ( R ) of each phagosome was measured using the “Measuring” tool provided by ImageJ , and V is calculated as 1 . 333 π R3 . Under the described image-capture conditions , the pixel distance is 0 . 133 μm/pixel . To measure the size of yolk droplets , the threshold tool was adjusted to mark all droplets labeled with YP170::GFP in an embryo . The size of every droplet larger than 0 . 13 μm2 was analyzed and collected to generate the distribution histogram .
The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank ) accession numbers for RAB-7 , CTNS-1 ( a ) , and CTNS-1 ( b ) are CAA91357 , CAA88102 , and CAD56564 , respectively . | Cells undergoing programmed cell death , or apoptosis , within an animal are swiftly engulfed by phagocytes and degraded inside phagosomes , vesicles in which the apoptotic cell is bounded by the engulfing cell's membrane . Little is known about how the degradation process is triggered and controlled . We studied the degradation of apoptotic cells during the development of the nematode Caenorhabditis elegans . Aided by a newly developed live-cell imaging technique , we identified multiple cellular events occurring on phagosomal surfaces and tracked the initiation signal to CED-1 , a phagocytic receptor known to recognize apoptotic cells and to initiate their engulfment . CED-1 activates DYN-1 , a large GTPase , which further activates downstream events , leading intracellular organelles such as endosomes and lysosomes to deliver to phagosomes various molecules essential for the degradation of apoptotic cells . As well as establishing a temporal order of events that lead to the degradation of apoptotic cells , the results suggest that phagocytic receptors , in addition to initiating phagocytosis , promote phagosome maturation through the sequential activation of multiple effector molecules . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"developmental",
"biology",
"cell",
"biology",
"genetics",
"and",
"genomics"
] | 2008 | Phagocytic Receptor CED-1 Initiates a Signaling Pathway for Degrading Engulfed Apoptotic Cells |
The progression and variation of pathology during infections can be due to components from both host or pathogen , and/or the interaction between them . The influence of host genetic variation on disease pathology during infections with trypanosomes has been well studied in recent years , but the role of parasite genetic variation has not been extensively studied . We have shown that there is parasite strain-specific variation in the level of splenomegaly and hepatomegaly in infected mice and used a forward genetic approach to identify the parasite loci that determine this variation . This approach allowed us to dissect and identify the parasite loci that determine the complex phenotypes induced by infection . Using the available trypanosome genetic map , a major quantitative trait locus ( QTL ) was identified on T . brucei chromosome 3 ( LOD = 7 . 2 ) that accounted for approximately two thirds of the variance observed in each of two correlated phenotypes , splenomegaly and hepatomegaly , in the infected mice ( named TbOrg1 ) . In addition , a second locus was identified that contributed to splenomegaly , hepatomegaly and reticulocytosis ( TbOrg2 ) . This is the first use of quantitative trait locus mapping in a diploid protozoan and shows that there are trypanosome genes that directly contribute to the progression of pathology during infections and , therefore , that parasite genetic variation can be a critical factor in disease outcome . The identification of parasite loci is a first step towards identifying the genes that are responsible for these important traits and shows the power of genetic analysis as a tool for dissecting complex quantitative phenotypic traits .
Forward genetics provides a powerful tool for analysing phenotypes and identifying genes that are responsible for a number of important traits . The importance of the linkage mapping approach is particularly appropriate for the analysis of phenotypes for which there is no obvious candidate gene , or when more than one gene is predicted to be involved in a phenotype , so called complex traits or quantitative trait loci . Such quantitative traits include disease severity , where there is a range of pathogenesis phenotypes caused by a particular pathogen . African trypanosomes cause a disease syndrome of high morbidity and mortality across large areas of sub-Saharan Africa in both humans ( sleeping sickness ) and domesticated animals ( Nagana ) . The parasites are spread by blood-feeding tsetse fly vectors ( Glossina ssp . ) , which inject the organisms into the mammalian host's bloodstream where they replicate extracellularly resulting in a chronic wasting condition . Two subspecies of Trypanosoma brucei , T . b . rhodesiense and T . b . gambiense , are able to infect humans causing a disease that is normally fatal and affects approximately 70 , 000 people per year [1] ( although this number is undoubtedly a gross underestimate [2] ) . In livestock , tens of millions of animals are affected each year with T . brucei and the related pathogens T . congolense and T . vivax [3] . In any host-pathogen relationship , variation in disease outcome can arise from differences between either hosts , pathogens , or both . In trypanosome biology , variation in parasite virulence has been well documented but the genetic basis for this has been largely unexplored . Classically , the two T . brucei subspecies have been described as causing different pathology; T . b . rhodesiense causes a short , acute disease , while that seen with T . b . gambiense is more chronic and less severe [4] . However , the clinical distinctions are unquestionably less clear than textbooks suggest . Laboratory experiments have demonstrated differences in pathogenicity during mouse infections with different strains of T . b . gambiense [5] , [6] , and recently it has been shown that there is markedly different pathology elicited in terms of disease severity and progression between T . b . rhodesiense patients in geographically distinct areas in Uganda and Malawi [7] , [8] . In the major cattle pathogens , T . congolense and T . vivax , variation between isolates in clinical pathology are also well documented [9] , [10] , [11] . Variation in parasite virulence phenotypes that can be directly attributed to trypanosome genetic variation has been experimentally demonstrated in mice in several studies using T . brucei and T . congolense [12] , [13] . Thus , variation is well known and clinically important . The genetic basis for this variation however , has not been investigated . This is in marked contrast to the very considerable effort that has been directed towards dissecting the genetic basis behind variation in pathology ( ‘trypanotolerance’ ) attributable to the mammalian host in response to infections with the major veterinary pathogen , T . congolense in both mice [14] , [15] and cattle [16] . These studies have resulted in the identification of loci in both host systems that contribute to the control of infection . As all three major African trypanosome species are responsible for a wide range of virulence phenotypes , understanding the genetic determinants of this variation is an essential factor to be integrated into any model of pathogenesis of trypanosomiasis [7] , [12] , [17] . Therefore , the contribution of the host to the control of disease and the parasite to the pathogenesis must both be examined in order to produce a holistic picture of host-parasite interactions and the survival of the host . Observed differences in virulence between trypanosome strains point to a genetic basis for the spectrum of disease caused in the host and therefore provide a route for identifying the parasite factors that cause disease in the mammalian host . The use of a classical genetic linkage mapping approach , in which the inheritance of phenotypic traits are analysed in progeny of genetic crosses and examined for co-segregation with genetic markers , is a valuable route to identifying genes and loci that has been used for a number traits in pathogenic parasites . This approach is dependent upon the development of a genetic map for each species . Recently , genetic maps have been generated for a number of haploid parasites , for example , Plasmodium falciparum [18] , Plasmodium chabaudi [19] , Toxoplasma gondii [20] , and Eimeria tenella [21] , opening up the possibility of using classical genetic analysis to identify genes involved in important phenotypes . For example , the identification of protective antigens that enable survival upon exposure to the host immune response in Eimeria species [22] and P . falciparum [23] , a gene that influences the ability to invade red blood cells in P . falciparum [24] , and genes that confer resistance to chloroquine and quinine in P . falciparum [25] , [26] . The haploid nature of these organisms means that the contribution of a single allele to a phenotype can be measured in isolation without the complication of an effect of a second allele , allowing for the identification of genes involved in various phenotypes via a linkage mapping approach with remarkably few progeny . Perhaps the most compelling case for the power of genetic analysis is that of parasite virulence in T . gondii , whereby Quantitative Trait Loci ( QTL ) associated with virulence were identified [20] , which directly led to the identification of secreted kinases as the key genes involved in strain specific pathogenesis of toxoplasmosis [27] , [28] , [29] . Indeed , the relevance of these genes has been confirmed in T . gondii field strains [30] and these latter studies elegantly underline the application of the classical genetic approach to identifying genes of large effect in parasites [31] , [32] . The development of genetic maps for two sub-species of T . brucei , T . b . brucei and T . b . gambiense [33] , [34] provides the potential for using a classical genetic approach in this organism , although the diploid nature of this parasite means that only phenotypes that are heterozygous in the parental strain for which the map was generated can be mapped using F1 progeny . Pathology in trypanosome-infected hosts is multi-faceted and we have chosen to focus on four indices of pathology; anaemia , reticulocytosis , splenomegaly and hepatomegaly . These traits are all important in both humans and cattle in infections with all three species of trypanosome [11] , [35] , [36] . Here , we report a genetic analysis of these inherited host pathology traits and show that two , splenomegaly and hepatomegaly , resulted in a highly significant QTL on chromosome 3 of the T . brucei genome .
The maintenance and care of experimental animals complied with the appropriate legislation; the UK Animals ( Scientific Procedures ) Act , 1986 , and with the national and University of Glasgow maintenance and care guidelines . A panel of 39 independent F1 T . brucei progeny clones had been previously generated from a genetic cross between a strain originally isolated from a tsetse fly ( TREU927 ) and a second isolated from a hartebeeste ( STIB247 ) [37] , [38] . Parental and progeny clones were previously genotyped for 182 microsatellite markers [33] and thirty one progeny were used in the current study ( Fig . 1 ) . The eight further clones isolated from the TREU927/STIB247 cross are only available at present as the insect life cycle stage procyclic forms and are thus not suitable for in vivo mammalian studies . The two parental lines were each transmitted through tsetse flies and recloned ( using methods described in [37] ) so that their transmission/passage history was very similar to that of the hybrid progeny . All parasites used readily infect rodents and passages in our laboratory are kept to a maximum of six times in mice from tsetse transmission , in order to maintain tsetse transmissibility and minimise selection by growth in rodents [39] . Five mini- and microsatellite markers , all on different chromosomes of T . brucei , were used for routine genotyping of all infections ( Ch1/MS42 , Ch2/PLC , Ch3/IJ15/1 , Ch5/JS2 and Ch9/49; for primer sequences and PCR conditions see [33] ) . Parasites were grown from stabilates ( cryopreserved in liquid nitrogen ) in a donor ICR mouse ( Harlan , UK ) . Parasites in logarithmic growth were harvested from the donor mouse by terminal exsanguination . The trypanosomes were counted in triplicate in an improved Neubauer haemocytometer , and diluted in Carter's Balanced Salt Solution ( CBSS ) to 1×104 trypanosomes per 0 . 2 ml inoculum . Parasites were then inoculated via the intraperitoneal route into experimental BALB/c mice ( Harlan , UK ) . Five mice were infected per strain , and an equivalent number of uninfected mice were included as controls . Infections with all of the progeny were completed in nine different batches and , in each case , infections with both parental strains were used to ensure consistency of phenotypes across all of the mouse batches , and to provide a means of minimising batch-to-batch variation ( see below ) . At day 10 post-infection , mice were euthanased . Day 10 was chosen as it was the time point at which the greatest number of pathogenesis phenotypes were significantly different between the parental strains [Morrison et al . , submitted] . We chose to measure two features that can be readily related to known clinical pathology in multiple host species – organomegaly and anaemia . At day 10 post-infection , mice were euthanased and the spleen and liver dissected and weighed . After weighing , the spleens were snap frozen in liquid nitrogen , and stored at −80°C . A FACs based assay was developed to measure anaemia in the mouse model [Morrison et al . , submitted] . Briefly , parameters for FACs analysis were determined for each cell population ( trypanosomes , red blood cells , reticulocyctes and white blood cells ) , by assaying each population in isolation with a known number of fluorescent beads , allowing the gating and calculation of cell populations from whole infected blood . For each strain 5 µl blood was collected and added to 2 µl of in CBSS containing 100 U/ml heparin , followed by the addition of 198 µl 1% paraformaldehyde in PBS pH 7 . 4 , mixed thoroughly to ensure no aggregates formed and fixed for 30 minutes at room temperature . To stain reticulocytes , 745 µl thiazole orange ( 100 ng/ml in PBS pH 7 . 4 ) was added to the blood/fixative mixture and allowed to stain for 1 hour in the dark at room temperature [40] . Fluorescent beads ( Countbright™ absolute counting beads for flow cytometry , Invitrogen , UK ) were prepared by diluting to 1000 beads/µl and ensuring even mixing by sonication . Fifty µl of the bead mixture was then added to the blood/thiazole orange preparation ( total volume 1 ml; 50 , 000 beads/ml ) . FACS was carried out using a Becton Dickinson FACSCalibur , using detector FL1-H for thiazole orange . Data were analyzed using CellQuest version 3 . 3 . For each sample 10 , 000 events were counted five times . This assay differs from that recently described by Antoine-Moussiaux et al [41] and is described in full in Morrison et al ( submitted ) . The data for all four phenotypes were normally distributed ( D'Agostino-Pearson normality test , p>0 . 05 in all cases ) , and this allowed calculation of Pairwise correlations of phenotype data using Pearson correlation coefficient and a two-tailed P-value , and from this a value of r2 was generated . For QTL analysis , the mean phenotype value for each progeny clone was calculated as a percentage of the phenotype for the TREU927 parental strain in that particular batch to take batch-to-batch variation into account . QTL analysis was performed using MapManager QTX software [42] for the T . brucei genetic map [33] . The variance was calculated from the five biological replicates of each phenotype for each progeny . The Likelihood Ratio Statistic ( LRS ) significance values were calculated by 1000 random permutations of the phenotypes relative to the genotypes that were intrinsic to the experimental data set ( to obtain the equivalent LOD score , the LRS is divided by approximately 4 . 6 ) . Thresholds of statistical significance ( χ2 statistic ) were calculated from each permutation test for each phenotype . The calculated threshold values were in line with those suggested for genome-wide scan studies [43] . For a simplified explanation of the inheritance model and relevant features of the TREU927 × STIB247 cross see Fig S1 . An additional genetic marker , ( TB3/19 ) , was developed for T . brucei chromosome 3 based on the allele specific amplification of a 1 kb fragment of DNA of the TREU927 parental strain and the failure to amplify the other TREU927 allele or STIB247 parental alleles , due to a single nucleotide polymorphism in the primer site ( s ) . As a positive control for DNA integrity , primers directed against a second non-polymorphic 1 kb fragment were used ( TB3/20 ) , which amplified both TREU927 alleles . This allele-specific PCR ( TB3/19 ) segregated in the progeny and was incorporated into the genetic map . Primer sequences were Ch3/19A TGAGTTCCTCTTGCACTCCC , Ch3/19B TCTTTACGTGTGCGCGCTAGG , Ch3/20A GATGTAATGTCCCTTCGGATTGCG and Ch3/20B GTGTCTTGTAGTTTATGACGGC; PCR conditions were described previously [33]; Genotyping gaps in the original map were also filled ( for version 2 of the genetic map , see http://tinyurl . com/trypmap ) .
The trypanosomes strains TREU927 and STIB247 consistently differ in the pathogenesis they induce , including the degree to which they cause anaemia , reticulocytosis , splenomegaly and hepatomegaly ( Fig 1 ) . We measured these parameters in nine batches of mice and consistently observed the same pattern with greater splenomegaly and reticulocytosis in 247-infected mice and marked anaemia in 927-infected mice ( Fig 1 and Fig S2 ) . Whilst the patterns were consistent , the absolute values showed some batch-to-batch variation and to account for this , subsequent data are presented as percentages of the 927 parental levels . A detailed analysis of hepatomegaly development over multiple time points permitted robust statistical analysis by 2-way ANOVA and showed hepatomegaly to be greater in STIB247-infected mice [Morrison et al , submitted] , although the data for day 10 show a smaller difference than was observed for the other three phenotypes ( Fig S2 ) . In order to investigate if there is a genetic basis for this variation in these phenotypes , experimental mouse infections were carried out with 31 independent F1 progeny from a genetic cross between TREU927 and STIB247 . The mean data for each phenotype for each progeny clone are illustrated in Figure 2 , revealing a continuous distribution for each phenotype . All four phenotypes show marked transgressive segregation with between 10 and 20 hybrids displaying average values outside the parental range . Interestingly , few values fall below that observed for TREU927 and a large number of the progeny display values far greater than either parent . This heterosis was particularly marked for splenomegaly , hepatomegaly and reticulocytosis but less so for red cell numbers . All four phenotypes show classical quantitative inheritance with no clear segregation into different classes in the progeny , but the progeny clones at the top and bottom ends of each distribution show clear and reproducible differences , suggesting that parasite derived genetic determinants contribute to each phenotype . Pairwise correlations were calculated for all phenotypes , revealing three combinations that show significant correlation ( Fig 3 ) . The most significant correlation was between red blood cell numbers and percentage reticulocytes , which suggested that 57% of the variance is shared between the two phenotypes . There were also significant correlations between liver and spleen weights ( 44% of shared variance ) and reticulocyte percentage and liver weight ( 21% ) . The relatively continuous distribution of phenotypic data from the F1 progeny in all cases suggests that multiple loci are involved in determining the phenotype . In order to investigate the genetic basis determining these traits a quantitative linkage mapping approach was therefore applied to the data using the available genetic map . The T . brucei genetic map created from a TREU927 × STIB247 cross [33] comprising 196 markers spread across the 11 megabase chromosomes ( but not covering subtelomeric regions , or mini – and intermediate chromosomes ) , was used to analyse whether the phenotypes ( spleen weight , liver weight , reticulocyte percentage and red blood cell numbers ) co-segregated with genetic markers , using QTL analysis .
We have used genetic linkage analysis to investigate a variety of pathogenesis phenotypes in infections with T . brucei . Our approach , using genetically distinct parasite strains in one inbred host background effectively means that observed differences in infection profile must originate from parasite genotype differences . Differences caused by host factors also exist , but have been minimised by the use of inbred mice and controlling for any batch-to-batch variation . Our results clearly show that variation in pathology has a parasite determined genetic basis and that using a linkage mapping approach it was possible to identify a region of the genome on chromosome 3 that has a major effect on the gross pathology in mice . This study is an important step , therefore , towards identifying the parasite virulence factors that underlie differential pathogenesis . The phenotype data in this study do not segregate in a manner that would indicate they are determined by a single gene , unlike that of previous studies in haploid parasites , where traits in progeny could definitively be scored as ( for example ) virulent , intermediate or non-virulent [20] , or drug resistant or sensitive [25] . The phenotypes we have examined are , or undoubtedly can be , the end product of multiple processes , and therefore will have multiple genes contributing to the variance . They are therefore true ‘quantitative traits’ , and for this reason QTL analysis was the most appropriate tool to search for linkage between trypanosome genotype and host infection phenotype . The identification of a highly significant QTL , despite the complex nature of the phenotypes , suggests that we have identified a parasite locus that determines a process responsible for a large proportion of the variance observed . The correlation of liver and spleen weights , and linkage of each to the same region of chromosome 3 , means that it is reasonable to assume that the same locus is responsible for the differential pathogenesis in both phenotypes . Whether this is due to a parasite gene product directly influencing both processes individually , or whether the gene product defines the polarisation or sequence in which multiple downstream host and parasite pathways proceed , is one of the questions that remain to be answered . Certainly the spleen and liver are significant organs with respect to the response to infection , and splenomegaly is one of the classical clinical signs observed in experimental cases of animal trypanosomiasis and also in natural cases of animal and human disease [35] , [44] , [45] . The spleen and liver obviously both have roles with respect to the innate immune response [46] and , with respect to trypanosome infections , the direction in which polarisation of the immune response proceeds seems to determine the progression of disease [47] , [48] , [49] , [50] . We have previously shown that in infections with the two parental strains , pathways of the innate immune response , macrophage polarisation ( IL10 , RXR/LXR pathways ) , and alternative macrophage activation are the most significantly differentially regulated [Morrison et al . , submitted] . The strain specific pathogenesis is therefore not only due to a difference in magnitude of the same pathways , but distinct pathways . It seems reasonable , therefore , to postulate that there is a parasite gene product that determines this polarisation , and the gene is located within the locus on chromosome 3 . The confidence interval of the major chromosome 3 QTL for splenomegaly and hepatomegaly ( TbOrg1 ) includes 383 genes . Of the genes within the QTL , 206 are conserved hypothetical proteins , 109 have function assigned by sequence homology , 62 have a predicted signal peptide , 10 are sequence orphans , 3 are pseudogenes and only 16 have been experimentally characterised ( Table S1 ) . Given the number of candidate genes , a fine-mapping approach is required to narrow down the region of linkage and reduce the number of genes to a level where candidates can be experimentally examined . There is scope for this approach , as the QTL interval for both phenotypes contains 12 crossovers in the 31 progeny , and therefore informative markers will further define the linkage boundaries . Fine mapping , in which more informative genetic markers within the QTL region are identified , can be undertaken in a number of ways , the most definitive of which will be to sequence the genomes of the parental strains to identify all single nucleotide polymorphisms within the locus , which can then be used to define the crossover boundaries in the progeny more precisely . There are additional and complementary techniques that will also aid in defining candidate genes , for example; the generation of further progeny in order to incorporate a greater number of informative crossovers in the genetic map and the identification of genes that are heterozygous for 927 ( by identifying heterozygous single nucleotide polymorphisms ( SNPs ) in the relevant chromosome sequence ) within the QTL ( given the necessity for the gene to be heterozygous to obtain segregation in this cross ) . Additionally , gene expression data could be used to rule out , for example , those genes only expressed in the procyclic life cycle stage , and inferences of gene function could also be made from the genome sequence data . Similar post-genomic approaches were used successfully to identify T . gondii virulence factors , when starting from QTLs spanning regions of a similar magnitude [27] . Additionally , while the approach using F1 progeny clearly allows the identification of dominant heterozygous loci ( Fig S1 ) , a limitation is that it will not detect co-dominant or recessive loci . Therefore , using the same approach , but with an F2 cross , would allow the dissection of further loci that are not detectable with the current approach , and provide a more comprehensive picture of the parasite genes contributing to pathogenesis . However , no successful F2 cross has been reported in trypanosomes thus far . The QTLs identified on chromosome 2 for splenomegaly , hepatomegaly and reticulocytosis ( TbOrg2 ) are potentially interesting . In all three cases there are multiple markers linked and the loci of the three phenotypes partially overlap , suggesting that the correlations between the traits ( two of which are highly significant and the third approaches significance ) may have some causal basis worthy of further investigation . In contrast , there are no QTLs in common for red blood cell numbers and reticulocytosis despite the very strong correlation between these two traits . One explanation for this is that the spleen , and to a lesser extent the liver , are haematopoietic organs , and therefore linked to reticulocyte production . In contrast , the change in red blood cell numbers will largely be a function of the erythrophagocytosis that occurs in early infection [36] . The implication is that , whilst the traits may be linked mechanistically , the parasite determinants that drive organomegaly/reticulocytosis and red blood cell numbers may be different . In summary , we have demonstrated for the first time that forward genetics provides a powerful tool to map genes in the trypanosome genome that are responsible for causing complex phenotypes in the mouse host . We have defined two loci , one a major locus on Chromosome 3 ( TbOrg1 ) that contributes to a significant amount of the variance observed in splenomegaly and hepatomagaly during mouse infections with T . brucei , and a second QTL on Chromosome 2 that contributes to splenomegaly , hepatomegaly and reticulocytosis in these infections ( TbOrg2 ) . Identification of the trypanosome genes that mediate the differential pathogenesis will provide a fundamental insight into the mechanisms by which the parasite causes disease in the mammalian host , and these findings may provide potential for developing therapeutic interventions to alleviate the disease associated with this pathogen . | Trypanosomes are single-celled organisms that are transmitted between animal hosts by the tsetse fly . These parasites infect a wide range of mammals and in sub-Saharan Africa are extensively debilitating to livestock , and some species are also able to infect humans causing a disease , sleeping sickness , that is usually fatal unless treated . Some trypanosome strains cause more severe disease than others , and studying these differences may allow the identification of how serious disease is caused . We approached this problem by looking at how differences in disease symptoms ( enlarged spleen and liver , and reduced blood cell numbers ) that are caused in infections in mice with two strains of Trypanosoma brucei , TREU927 and STIB247 . These disease manifestations are clinically relevant in human and livestock trypanosome infections . Examining how the symptoms are inherited in infections with offspring of a cross between the two strains allowed the identification of a region of the T . brucei genome that contains a gene ( or several genes ) that contributes significantly towards the enlarged spleen and liver observed in infected mice . This is a first step towards identifying the parasite genes that cause disease in the host ( virulence factors ) , which may provide routes for developing novel therapies against the disease . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"genetics",
"and",
"genomics/complex",
"traits"
] | 2009 | A Major Genetic Locus in Trypanosoma brucei Is a Determinant of Host Pathology |
In spite of its evolutionary significance and conservation importance , the population structure of the common chimpanzee , Pan troglodytes , is still poorly understood . An issue of particular controversy is whether the proposed fourth subspecies of chimpanzee , Pan troglodytes ellioti , from parts of Nigeria and Cameroon , is genetically distinct . Although modern high-throughput SNP genotyping has had a major impact on our understanding of human population structure and demographic history , its application to ecological , demographic , or conservation questions in non-human species has been extremely limited . Here we apply these tools to chimpanzee population structure , using ∼700 autosomal SNPs derived from chimpanzee genomic data and a further ∼100 SNPs from targeted re-sequencing . We demonstrate conclusively the existence of P . t . ellioti as a genetically distinct subgroup . We show that there is clear differentiation between the verus , troglodytes , and ellioti populations at the SNP and haplotype level , on a scale that is greater than that separating continental human populations . Further , we show that only a small set of SNPs ( 10–20 ) is needed to successfully assign individuals to these populations . Tellingly , use of only mitochondrial DNA variation to classify individuals is erroneous in 4 of 54 cases , reinforcing the dangers of basing demographic inference on a single locus and implying that the demographic history of the species is more complicated than that suggested analyses based solely on mtDNA . In this study we demonstrate the feasibility of developing economical and robust tests of individual chimpanzee origin as well as in-depth studies of population structure . These findings have important implications for conservation strategies and our understanding of the evolution of chimpanzees . They also act as a proof-of-principle for the use of cheap high-throughput genomic methods for ecological questions .
The history and population structure of the common chimpanzee , Pan troglodytes , are incompletely understood . Traditionally , three subspecies have been described: the western chimpanzee ( P . t . verus ) , central chimpanzee ( P . t . troglodytes ) and eastern chimpanzee ( P . t . schweinfurthii ) . Analysis of mitochondrial DNA ( mtDNA ) variation led to the proposal of a fourth , “Nigerian” chimpanzee subspecies ( P . t . vellerosus , since renamed P . t . ellioti [1] ) as a sister taxon to P . t . verus occurring in an area of Nigeria and Cameroon east of the Niger river and north of the Sanaga river ( Figure 1 ) [2] , [3] . This new subspecies has been recognized by many taxonomists and conservation biologists [4] , [5] . Subsequent analyses of autosomal microsatellite data , in one case based on few loci [6] , and in another including few individuals designated a priori as P . t . ellioti [7] , found little evidence to distinguish P . t . ellioti from P . t . troglodytes , which is distributed south of the Sanaga river ( Figure 1 ) . Very recently however a microsatellite-based study of 94 individuals with 27 loci [8] has established that up to five groups of common chimpanzees , including P . t . ellioti , can be distinguished genetically . In this study we provide a complementary analysis using very different data and analytical methodology that allows a direct comparison with human data . For most animals , the definition of a subspecies as “a collection of populations occupying a distinct breeding range and diagnosably distinct from other populations” [9] would be uncontroversial . However , our close evolutionary relationship with chimpanzees , and the parallels that can be drawn between chimpanzees and humans , makes this terminology increasingly uncomfortable , and in some cases controversial , and so we prefer to avoid it . Whatever term is used , modern genetic methods clearly have the potential to make the assessment of distinctiveness more objective and precise than in the past and it should now be possible to confirm or refine earlier judgments that were based on other criteria or limited data . The development of modern high-throughput SNP genotyping technologies has revolutionized many aspects of human genetics , including our understanding of the history and demography of human populations [10]–[13] . To date , the impact of such methods in non-human species has been limited ( e . g . [14] , [15] ) . Here we apply these technologies to chimpanzees , and show that they can clearly resolve the genetic distinctness of P . t . ellioti , and that , for conservation purposes , small subsets of SNPs can be used to distinguish previously recognized populations . Our major source of SNPs was those arising from sequencing reads of a single individual ( “Clint” ) from the chimpanzee genome project [16] . A notable finding is that , in spite of the severe ascertainment biases inherent in this SNP discovery ( largely a single individual , from only one of the populations ) , analyses based on the resulting SNPs remain powerful , suggesting that the same may be true in other species for which there have been genome projects . We also demonstrate the potential benefits of haplotype-based analyses in combination with genomic SNP data in defining and quantifying population relationships .
To address the question of whether Pan t . ellioti is genetically distinct from other populations , we obtained DNA samples from Cameroonian chimpanzees which we analysed along with samples from captive Western ( verus ) and troglodytes chimpanzees . Eastern chimpanzees ( P . t . schweinfurthii ) , with their distinct geographical distribution , were not sampled in the current study . We sequenced 12 autosomal fragments of ∼1 kb and genotyped 691 SNPs from 22 autosomal regions of 40–80 kb [17] in order to resolve genome-wide relationships , and compared the results with inference from the mitochondrial HV-I locus . We applied a number of different methods to the analysis of these data to assess the relationships and genetic clustering amongst the sampled individuals . The first set of methods ( principal components and Structure ) were based on the marginal data at each genotyped SNP . We then calculated FST from the DNA sequence data , and finally applied recently developed methods which exploited information on the joint distribution of SNP alleles within haplotypes . Using the first two principal components of the data from all 818 SNPs , 52 of the 54 chimpanzees studied clustered into three distinct , non-overlapping groups ( Figure 2a ) . These clusters are consistent with three genetically distinct populations represented amongst the study chimpanzees: captive Western ( P . t . verus ) chimpanzees form one cluster while Cameroonian chimpanzees are divided into two genetically distinct clusters , one of which we infer to correspond to P . t . ellioti , whose existence had been the subject of uncertainty . We note that two individuals in the P . t . ellioti cluster had previously been designated P . t . troglodytes based on mtDNA sequence , a point to which we return below . Two individuals ( C024 , C025 ) with P . t . troglodytes-like mtDNA lie between the presumptive P . t . verus and P . t . troglodytes clusters , and records have subsequently revealed that these are indeed first-generation hybrids produced in captivity . A similar conclusion comes from a different perspective when the software Structure [18] , [19] is used to estimate the proportion of each individual's genome that comes from each of several ancestral populations . With k = 3 presumptive populations , the same three groups were recovered cleanly with little estimated admixture except for the two hybrids ( Figure 3 ) , and where there was evidence for co-ancestry , it was detected between the ellioti and troglodytes groups , rather than involving verus chimpanzees . This suggests more recent interaction between P . t . ellioti and P . t . troglodytes than either has had with P . t . verus , although an effect of SNP ascertainment could not be ruled out . We note that the model underlying Structure assumes no linkage disequilibrium between loci , whereas our data do exhibit such correlations because of the clustering of SNPs . The expected effect of this in the Structure model is an over-estimation of precision , rather than bias [19] , but nonetheless our Structure analysis should be interpreted with some caution . Next , we calculated pairwise FST , a commonly-used measure of the proportion of total genetic variation occurring between populations . Potential confounding effects from SNP ascertainment complicate interpretation of FST values calculated from the genotype data , so we restricted these analyses to our re-sequencing data alone ( 104 of 818 SNPs , also eliminating 3 sequenced loci showing evidence of positive selection ) [20] . Consistent with Structure's view of relative amounts of co-ancestry , FST between P . t . ellioti and P . t . troglodytes ( 0 . 134 , 95% CI 0 . 105–0 . 162 ) is slightly lower than , but cannot be formally distinguished from , that between P . t . troglodytes and P . t . verus ( 0 . 177 , 95% CI 0 . 129–0 . 225 ) or between P . t . ellioti and P . t . verus ( 0 . 190 , 95% CI 0 . 145–0 . 235 ) . The troglodytes – verus figure in our data is lower than the 0 . 29 for Central vs . Western chimpanzees previously estimated from re-sequencing data [21] , presumably due to sampling differences ( either of loci or individuals ) between the two studies . When genetic data is collected from tightly linked variable sites , exploiting patterns of non-random association ( i . e . linkage disequilibrium ) can increase power to identify population structure over single-SNP analyses [22] , [23] . Informally , haplotype-based approaches have many of the advantages in terms of discriminatory power of other multi-allelic systems such as microsatellites , but in addition , our understanding of the evolutionary mechanisms involved means that there is a natural sense of the evolutionary distance between haplotypes . Sensible haplotype-based analyses can thus be more powerful than SNP-based approaches in using considerably more genetic information in comparing individuals , and in our context can thus be informative about differentiation at timescales shorter than those over which drift can be detected in SNP frequency differences . Additionally , haplotype-based analyses may be less susceptible to biases in SNP discovery [22] . Conversely , while haplotype-based methods can increase power to detect population structure , statistical methodology to fit explicit models of isolation , migration and fluctuating population size [24] to such data is so far lacking . We analysed similarities in patterns of haplotype variation among individuals for the 691 clustered autosomal SNPs using a so-called copying model applied to estimated haplotypes from each individual [25] , [26] . In effect , for each small chromosomal segment in one of the haplotypes of a particular individual , the approach looks amongst the haplotypes of the other sampled individuals to find the one with which it is most closely related , in the sense of most recently sharing a common ancestor . This is done under a model in which shared ancestry is likely to be the same for chromosomal segments which are very near to each other ( in terms of genetic distance ) . The primary results of such an analysis are estimates of the most recent shared ancestry across each locus in each haplotype . For a particular chimpanzee , these can be aggregated to calculate the estimated proportion of the sampled regions for which it is most closely related to each of the other chimpanzees . These estimates are shown in Figure 4a . The figure provides a visual summary of the patterns of most-recently-shared ancestry within and between the three population groups . In a randomly mating population , the haplotypes in a particular individual will share similarities with many others across the sample , while in the presence of population structure haplotypes will tend to be more similar to those of other individuals within the same population than to those in other populations . Figure 4b ( see also Table 1 ) provides a higher-level summary which aggregates information across populations to show , for each chimpanzee , the proportion of its sampled regions for which the most closely related haplotype comes from each of the three populations . Strikingly , Figure 4a and 4b show that across most of the sampled regions in each individual , the most closely related haplotype comes from the same population; in other words that the three populations are genetically quite distinct . This effect is most marked for the P . t . verus individuals , for whom the most closely related haplotype is virtually always in the same population . Haplotypes of P . t . ellioti and P . t . troglodytes chimpanzees respectively are typically most similar to those of other individuals within the same population , but occasionally to those of individuals from the other ( P . t . troglodytes and P . t . ellioti respectively ) population . The two previously noted hybrid individuals are clearly identified , and in addition it emerges that two of the P . t . ellioti chimpanzees had a higher level of shared ancestry than the other chimpanzees . The qualitative conclusions from the haplotype-based analysis thus mimic those from principal components and Structure , although reassuringly they explicitly model the correlations between nearby SNPs , in contrast to Structure . By applying the haplotype-based copying model to human data , we can compare quantitatively the extent of differentiation between the three chimpanzee groups with that between various human populations . Importantly , such analyses can allow for ascertainment effects . We show the copy model results for human data from the Phase II HapMap ( Frazer et al . 2007 ) in Figure 4c and 4d , comparing sampled individuals of European ( CEPH ) , African ( Yoruba , YRI ) and East-Asian ( Han Chinese , CHB ) descent in an analysis in which SNPs in the human data were re-ascertained to match characteristics of the chimpanzee data ( see “Data Analysis” ) . The average within- vs . between-population copying frequencies , that is , frequencies for the most-closely-related-haplotype , in these analyses are summarized in Table 1 . Levels of between-population similarity among the chimpanzee populations are lower than among the HapMap populations , suggesting that the chimpanzee populations are more distinct than even continental human populations . To test the robustness of this conclusion to choice of comparison data , we re-sampled Phase II HapMap individuals , genomic regions and ascertained SNPs , 100 times . Only three times was the level of within-population copying of a pair of human populations greater than that between any chimp population ( estimated within-population copying in each of Africa and East-Asia was greater than the estimated within-population copying in P . t . troglodytes for 3 of 100 re-samples ) . In Figures S1 and S2 , we colour fragments of chromosomes according to their assigned population of origin under the copying model , illustrating that the probabilities with which individual chimpanzee chromosome segments are assigned to specific populations are also higher than for human data . An equivalent analysis of the HapMap III African populations [27] showed that these African human populations are considerably less structured than the chimpanzee populations ( Figures S3 and S4 ) , as might be expected given the observation above that the chimpanzee populations are more differentiated even than continental human populations . Note that our comparisons with the human population samples are based on similar amounts of data as in our chimpanzee samples . With larger SNP datasets , the power to separate the human populations increases . We note that while it is theoretically possible to use the lengths of copied fragments in the copying model to estimate the timescale over which differentiation has occurred , our data is not well-suited to this because the shortness of the assayed regions means that relatively few breakpoints are observed , providing little information about the times of events in the history of chimpanzee populations .
We have applied a number of different analytical methods to an extensive set of SNP data from 54 chimpanzees . All of the methods point clearly to the existence of three distinct population groups , corresponding to three of the previously-described “subspecies” of chimpanzee P . t . verus , P . t . troglodytes , and P . t . ellioti , with the latter two groups sharing somewhat more similarity with each other than either does with P . t . verus . P . t . troglodytes and P . t . verus are two securely defined populations estimated to have diverged 0 . 4–0 . 6 million years ago [7] , [8] , [28]–[30] . Our analyses show P . t . ellioti to be clearly distinct from P . t . troglodytes with both groups equally distinct from P . t . verus , so that whatever terminology ( “population” or “subspecies” ) is applied to verus and troglodytes should equally be applied to ellioti . By way of comparison , we have shown that these three chimpanzee populations are more differentiated than even continental human populations , and also that in spite of the relatively close geographic proximity of the groups , particularly troglodytes and ellioti , the chimpanzee populations are considerably more distinct than the African populations sampled in HapMap III , suggesting rather differing demographic histories for the two sister species . In order to compare population comparisons based on the copying model with those based on more traditional FST approaches , we also calculated pairwise FST values for each of the 100 resamples of individuals and SNPs in our analyses of the three continental population samples . The results are summarized in Table 2 . We note that while the average values of pairwise FST across the 100 samples show the same pattern as copying proportions in the copying model , the sample-to-sample variation is larger . For example , the FST intervals for the central 95% of resamples for Europe-East Asia overlap those of Africa-Europe and Africa-East Asia , and for example for five of the 100 resamples the pairwise FST between Africa and Europe was actually smaller than that between Europe and East-Asia . In contrast , for the copying model analysis the 95% intervals for the proportion that Europe and East Asia copy from each other do not overlap with the 95% intervals for either copying from Africa , and the proportion that Europe copied from Africa was lower than the proportion Europe copied from East Asia in each of the 100 re-samples . This accurately reflects the fact that on average East Asia and Europe share more recent ancestry with each other than with Africa . One weakness of our study ( and some others ) is that we do not have definitive information on the geographic origin of all of the chimpanzees we have studied . All our analyses point to two very distinct population groups for the chimpanzees originating from eastern Nigeria and Cameroon . In the light of other genetic evidence for distinctiveness of individuals sampled from either side of the Sanaga River [3] , [8] , our assignment of one of our sampled groups as troglodytes and one as ellioti seems reasonable . Whilst our data alone could not rule out two distinct populations , one or both of which extends across the Sanaga River , this seems a priori unlikely – the river provides a natural barrier between the distinct populations , whereas if both were to exist on the same side of the river there seems no reason for their reproductive isolation—and at variance to other available evidence . Notwithstanding our lack of complete geographical information on sampled chimpanzees , the clear separation between all three populations , relative to the similarities within the populations , seems hard to reconcile with the suggestion that chimpanzee genetic variation is distributed more or less continuously across the species range ( cf [21] ) . The initial genetic description of P . t . ellioti was based on mtDNA sequence analysis [2] , [3] , which places most chimpanzees from parts of Nigeria and Cameroon north of the Sanaga river in a group sharing a common ancestor with P . t . verus , to the exclusion of P . t . troglodytes , a description made more robust by a recent analysis of complete mitochondrial genomes [31] , [32] . We compared the classification based on mtDNA with our genome-wide analysis and found that it classified 50 of 52 non-hybrid individuals correctly . Chimpanzees C127 and C541 had troglodytes-like mtDNA but ellioti autosomal SNP genotypes . ( The two known hybrid chimpanzees C024 and C025 had troglodytes-like mtDNA but were detectably intermediate in autosomal genotype ) . Thus the two systems generally agree , but , not surprisingly , single-locus mtDNA data is less reliable for classification than genome-wide data . The mtDNA-based picture of demographic relationships suggests that P . t . verus and P . t . ellioti are sister taxa [3] , [31] . Our data suggests this to be misleading , in two different respects . Firstly , as noted above , two individuals who are clearly P . t . ellioti , on the basis of extensive autosomal data , have mtDNA which clusters with P . t . troglodytes . Thus , mtDNA from ellioti individuals does not fall into a single clade on a mtDNA tree . If mtDNA is used both to classify individuals and to estimate trees for the resulting groups , there is always a danger , as seems to have occurred in this instance , that misclassification of individuals will lead to a simpler-looking tree than is actually the case . Secondly , the suggestion from the mtDNA data that ( many , but as noted above , not all ) ellioti individuals have mtDNA types which are closer to verus than to troglodytes individuals is strikingly different from the results of our analyses based on many independent autosomal loci , which places P . t . ellioti clearly closer to P . t . troglodytes than to P . t . verus . It is interesting to note that a study of morphological variation agreed with the picture obtained from autosomal loci [4] . Taken together , the mtDNA and autosomal results are difficult to reconcile with a simple demographic scenario based on population splitting , and suggest a more complex demographic history for the three populations we have studied , possibly including sex-biased gene flow . For many conservation applications , it would be desirable to be able to assign or classify individuals to populations based on a small number of loci . We developed and applied a method for choosing subsets of SNPs for classification based on their contribution to assignment probabilities ( see Methods ) . To avoid over-fitting , we divided our data set in two . A training dataset comprising half the samples from each population ( 27 of the 52 non-hybrid individuals ) was used to select informative SNPs for classification , with the other half of the individuals forming a test dataset in which the ability of the chosen SNPs to accurately classify individuals to populations was measured . For our data , we could essentially reproduce the discrimination obtained with the complete dataset of 818 SNPs with as few as 8 carefully selected SNPs in distinct regions of the genome ( Figure 2b ) . While there is still some danger of over-fitting from our relatively small sample sizes , we conclude that a small , well-chosen panel of probably 10–20 SNPs , assayed via either a set of PCR-based single-locus assays or a single multiplex SNP assay for forensic and conservation work , would be capable of analysing and classifying limited DNA samples at low cost . The exact size of panel used would depend on the requirement to identify individuals of mixed ancestry . This is particularly encouraging considering the extreme ascertainment bias inherent in our genotyped SNPs: for the chimpanzee , dbSNP at the time of our SNP selection reflected the composition of the chimpanzee draft genome , in which ∼91% of sequence traces came from a single P . t . verus individual ( ‘Clint’ ) , a further 4% from four other verus , and less than 5% from three P . t . troglodytes [16] . Notwithstanding this bias , 12 of our SNPs have an estimated allele frequency difference of >0 . 5 between ellioti and pooled troglodytes and verus chimpanzees . Our study thus confirms the utility of genomic resources even when ascertainment is sub-optimal . The confirmation of P . t . ellioti as a genetically distinct population of chimpanzee strongly supports efforts to treat this population as a separate management unit for conservation [33] This is of particular importance since while all chimpanzees are considered to be endangered [34] , P . t . ellioti , with an estimated 6 , 500 individuals remaining , is the least numerous population . In conclusion , using genomic resources we have assembled the largest SNP-based dataset for investigating chimpanzee population structure . It resolves an outstanding controversy in clearly establishing the fourth putative subspecies , Pan troglodytes ellioti , as a genetically distinct group . More generally , our results confirm the utility of high throughput SNP typing for evolutionary genetic and conservation analysis . However , we recognize that a full appraisal of chimpanzee population structure would require denser sampling from all four populations in addition potentially to comparative studies across primates that go beyond great apes and humans .
Blood samples were obtained from 35 wild-born orphaned chimpanzees of unknown geographic origin within Cameroon . Genomic DNA , extracted using standard procedures , was amplified ( GenomiPhi , GE Healthcare ) before genotyping . DNA samples were also obtained from 15 P . t . verus ( from Sierra Leone ) and 4 putative P . t . troglodytes ( unknown geographic origin ) chimpanzees held at the Biomedical Primate Research Centre in the Netherlands ( Table S1 ) . For chimpanzees in the Netherlands , all blood sampling was done in accordance with a protocol that was approved by the Institutional Animal Care and User Committee ( IACUC ) of the Biomedical Primate Research Center ( BPRC ) . For chimpanzees in Cameroon , blood samples were taken from orphaned individuals for haematological analysis as part of veterinary health screens . Mitochondrial HV-I fragments of 534 bp and fragments of ∼1 kbp from the genes CCR5 , SDF , CXCR4 , CX3CR1 , RANTES , CCR2 , SEC22L3 , ZNF445 , PTPN23 , CCRL2 , MC1R and HBB ( Table S2 , Table S3 ) were amplified by PCR and sequenced directly . PCR products with heterozygous indels were cloned and 10 clones were sequenced for each sample . For pairwise Fst analyses , 3 loci with evidence for directional selection ( CCR5 , CXCR4 and CX3CR1; 23 SNPs , MacFie et al . 2009 ) were removed from the analysis . A panel of 768 SNPs was designed for the GoldenGate Genotyping Assay ( Illumina , San Diego ) , using polymorphism information from the Chimpanzee Genome Project [16] via dbSNP v26 [http://www . ncbi . nlm . nih . gov/projects/SNP/] . The SNPs , arranged in 22 clusters of size 40–80 kbp on several autosomes , were screened using BLAST to ensure unique context . The panel has also been used to assess recombination rates in the 22 regions , orthologous to recombination hotspots in humans [17] . Across 54 samples , 58 SNPs failed visual inspection , 14 gave at least one no-call and 5 SNPs departed strongly from Hardy-Weinberg equilibrium within a population ( as initially labelled ) , leaving 691 SNPs for analysis . Population structure was assessed by pairwise FST in Arlequin ( with 95% CIs estimated by jackknifing ) [35] , PCA and SNP selection for assignment in the R Package [36] , and with Structure [18] , [19] , using the admixture model of ancestry , with correlated allele frequencies , run with a ‘burn-in’ of 100 , 000 iterations followed by a further 1 , 000 , 000 iterations . This model is not strictly applicable to data from sites in linkage disequilibrium , so this analysis is indicative only . SNPs were chosen for classification as follows: for each SNP a sample was assigned to the population in which its genotype was most probable , the 818 SNPs were ranked by their ability to classify the training samples and the best SNPs , from distinct loci , were chosen ( Table S4 ) . For the haplotype-based analysis , we inferred haplotypes and population-scaled recombination rates between adjacent SNPs using PHASEv2 . 1 . 1 [25] , [37]–[39] with ten times the default number of MCMC iterations . We then applied the Li and Stephens ( 2003 ) copying model to the inferred “best-guess” haplotypes as described in [40] but fixing the PHASE recombination rate estimates , inferring the expected number of haplotype segments that each chimp copies from every other chimp via 100 iterations of an Expectation-Maximization ( EM ) algorithm and precluding copying from the other haplotype within the same individual . Figures S1 , S2 , and S3 are based on 100 samples from the model using the converged E-M values . For comparisons with human data , we matched features of the chimpanzee dataset by randomly selecting 18 individuals per population using HapMap Phase 2 Release 21 or HapMap Phase 3 Release 2 consensus haplotypes . For each analysis , we then randomly selected 22 autosomal genomic regions , randomly selecting SNPs to match the SNP density and minor allele frequency distribution ( in bins of ( 0 . 0 , 0 . 1] , ( 0 . 1 , 0 . 2] , ( 0 . 2 , 0 . 3] , ( 0 . 3 , 0 . 4] , ( 0 . 4 , 0 . 5] ) for the respective 22 chimp regions . We ran the copying model using fixed genetic map estimates ( build 35 estimates for HapMap2 populations and build 36 estimates for HapMap3 populations ) scaled by an effective population size value of 30000 , the value that maximized the expected log-likelihood over a fixed grid of ( 10K , 20K , 30K , 40K , 60K , 300K , 25000K ) , though we note that results were similar for all scaling factors we considered . Ascertaining SNPs on a single randomly selected HapMap Phase2 CEPH individual or HapMap Phase3 Luhya ( Kenya ) individual not included in the sample gave similar results to those presented . Pairwise FST for each re-sample was calculated using the approach described in [41] . | Chimpanzees are viewed with fondness as our closest animal relatives and are valued by scientists for the biological and evolutionary insights they provide . In spite of this , the relationships between different populations of common chimpanzees are still relatively poorly understood , a situation that potentially threatens conservation efforts . Here we have used information gathered in the Chimpanzee Genome Project to design comprehensive tests of genetic variability that show unambiguously the existence of four genetically distinct groups ( or populations ) of common chimpanzee . We demonstrate that previous methods based on mitochondrial DNA sequences alone are not always accurate and show the feasibility of cheap new genetic tests of individuals' origins that could play an important role in conservation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"ecology",
"genetics",
"population",
"genetics",
"biology",
"genomics",
"population",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | Genomic Tools for Evolution and Conservation in the Chimpanzee: Pan troglodytes ellioti Is a Genetically Distinct Population |
Few live attenuated vaccines protect against multiple serotypes of bacterial pathogen because host serotype-specific immune responses are limited to the serotype present in the vaccine strain . Here , immunization with a mutant of Shigella flexneri 2a protected guinea pigs against subsequent infection by S . dysenteriae type 1 and S . sonnei strains . This deletion mutant lacked the RNA-binding protein Hfq leading to increased expression of the type III secretion system via loss of regulation , resulting in attenuation of cell viability through repression of stress response sigma factors . Such increased antigen production and simultaneous attenuation were expected to elicit protective immunity against Shigella strains of heterologous serotypes . Thus , the vaccine potential of this mutant was tested in two guinea pig models of shigellosis . Animals vaccinated in the left eye showed fewer symptoms upon subsequent challenge via the right eye , and even survived subsequent intestinal challenge . In addition , oral vaccination effectively induced production of immunoglobulins without severe side effects , again protecting all animals against subsequent intestinal challenge with S . dysenteriae type 1 or S . sonnei strains . Antibodies against common virulence proteins and the O-antigen of S . flexneri 2a were detected by immunofluorescence microscopy . Reaction of antibodies with various strains , including enteroinvasive Escherichia coli , suggested that common virulence proteins induced protective immunity against a range of serotypes . Therefore , vaccination is expected to cover not only the most prevalent serotypes of S . sonnei and S . flexneri 2a , but also various Shigella strains , including S . dysenteriae type 1 , which produces Shiga toxin .
Shigellosis is common worldwide . It is estimated that 164 . 7 million people are infected annually , resulting in 1 . 1 million deaths . About 70% of episodes and 60% of deaths involve children under 5 years-of-age [1] . In addition , growing antibiotic resistance [2] is a serious problem in all countries; this is compounded by the fact that so many people travel . Therefore , vaccines against shigellosis are being developed [3 , 4] . Shigella strains comprise four subspecies: S . dysenteriae , S . flexneri , S . sonnei , and S . boydii . These are further divided into 50 distinct serotypes [1 , 3] according to the immunogenicity of capsular lipopolysaccharide O-antigens . The World Health Organization has set Shigella dysenteriae type 1 ( Sd1 ) as a primary target for control because this strain produces Shiga toxin , a neuro-cytotoxic agent that causes hemolytic uremic syndrome ( HUS ) [5] . At present , S . sonnei is the most prevalent strain in industrialized countries [1]; however , imported epidemics of Sd1 with HUS have been reported [6] . A Global Enteric Multicenter study also indicated that S . flexneri ( 65 . 9% ) , particularly serotype 2a ( 20 . 2% ) , and S . sonnei ( 23 . 7% ) are the most prevalent strains isolated from patients aged <60 months at four sites in Africa and three sites in Southeast Asia [7] . The pathogenesis of Shigella strains is dependent on virulence plasmids [8] encoding common virulence factors [8 , 9] belonging to the type III secretion system ( T3SS ) [10] . The T3SS is a needle-like transporter complex expressed on the surface of bacteria [11]; the complex injects effector molecules , such as IpaBCDA proteins ( All genes and proteins are listed in Table 1 ) , into host cells to facilitate bacterial invasion . After propagation within the colonic epithelium , the bacteria spread via a mechanism involving the outer membrane protein VirG ( IcsA ) [12 , 13] . Many studies , including a prospective epidemiological surveillance study of a cohort of children in an endemic area , indicate that acquired immunity to shigellosis is O-antigen specific [14–16]; therefore , a majority of vaccine candidates have been developed to provide serotype-specific protection . The strategy was similar to that used to develop the practical pneumococcal and Hib vaccines , although immunization with inactivated cells or cellular components appeared insufficient , suggesting that some processes during the infection cycle are required for effective antigen presentation to the intestinal system . Such candidate Shigella vaccines comprise attenuated strains derived by mutation of virG and some metabolic genes [17–19] . However , a licensed vaccine is still not available because many candidates have failed to maintain a subtle balance between immunogenicity and safety [3 , 4] . Few studies support “cross protection” against both homologous and heterologous serotypes of Shigella strain . However , early field studies suggested the possibility . An attenuated S . flexneri 2a strain T32–Istrati was developed in Romania by serial passage of the culture . Oral administration of an extreme dose ( 0 . 5–2 . 0×1011 colony forming units ( cfu ) ) of live bacteria ) to volunteers was well tolerated . A field study of 32 , 000 children and 500 adults conducted in 1976–1980 reported 81% protection against S . flexneri 2a , 89% protection against S . sonnei , and 88% protection against other Shigella species [20 , 21] . Also , a Chinese study using the same strain ( the study enrolled 5000 vaccinees and 5000 controls ) reported 85% protection against homologous serotypes and 72% against heterologous serotypes [21 , 22] . However , no rationale for the observed cross protection was provided and no follow-up studies have been reported . Recent attempts to utilize common virulence proteins for immunization have also been reported . Intra-nasal immunization of a mouse model of pneumonia with recombinant T3SS effectors IpaB and IpaD along with a variant of a heat-labile toxin from E . coli ( dmLT ) protected against subsequent challenge with S . sonnei and S . flexneri [23] . Kim et al . showed that the C-terminal region of the outer membrane protein IcsP is common to all Shigella serotypes . Nasal immunization of a mouse model of pneumonia with a vaccine containing the C-terminal peptide of IcsP and dmLT provided cross protection ( >60% ) against S . flexneri 2a , S . flexneri 6 , and Sd1 [24] . Here , we report cross protection provided by a S . flexneri 2a-based vaccine candidate . The idea originated from our basic studies on regulation of T3SS expression . A mutant of S . sonnei harboring a deletion of the hfq gene encoding an RNA-binding protein lost the temperature- and osmotic-dependent regulation , showed increased production of T3SS and increased invasion into Hela cells . This occurred because expression of a T3SS regulator , invE ( virB ) [25 , 26] , was up-regulated via loss of regulation at the post-transcriptional level [27–29] . As observed for many other pathogens , loss of hfq results in attenuation [29] via repression of stress response regulators such as rpoE [30] and rpoS [31] . Attenuation via hfq mutation is also reported for an experimental vaccine against Salmonella typhimurium , but the study showed only homologous protection [32] . Increased antigen expression and simultaneous attenuation are expected to elicit protective immunity against Shigella strains of heterologous serotypes . Thus , we also characterized the hfq mutant by examining its potential effects as a vaccine . We performed two different challenge experiments to evaluate whether the Δhfq mutant provided protection against subsequent infection by Shigella strains of heterologous serotypes . In addition , we examined production of antibodies against various heterologous serotypes .
All experimental protocols were approved by the Animal Ethical Committees at the National Institute of Infectious Diseases ( NIID ) ( No . 208123 and 209002 ) and National Institute of Cholera and Enteric Diseases ( NICED ) ( No . Apo/80/06/05/2011 ) and conducted in accordance with Guidelines for the Proper Conduct of Animal Experiments ( Scientific Council of Japan ) and Guidelines for the Care and Use of Animals in Scientific Research ( Indian National Science Academy ) . Data are presented as the mean standard deviation ( SD ) . Statistically significant differences between individual groups were analyzed using an unpaired Student’s t-test . p<0 . 05 was considered significant . Experiments were performed at the two facilities . ( Fig 1E ) . For NIID , male Hartley guinea pigs ( 250–300 g ) were purchased from SLC Japan ( Fig 2 , S2 and S3 Figs ) , and S . sonnei ( HW383 ) and S . flexneri 1b ( 9268N ) were used for the challenge experiments . Sd1 ( TSH1669 and MD506 ) , EIEC ( NIID1 ) , and S . flexneri 1b ( 9268N and 9268N17-1 ) , 3a ( GTC-01924 ) , 6 ( GTC-01927 ) were used for microscopy . For NICED , male non-albino “Old-English-colored” guinea pigs ( 650–750 g ) were used for the colon loop experiments ( Fig 1 ) and for all subsequent experiments ( Figs 3–5 and S4 Fig ) . Sd1 ( NT4907 ) and S . sonnei ( IDH00968 ) were used for the challenge experiments . Plasmid pACYC-ipaBCDA was constructed by inserting T3SS effector genes ( identical to nucleotide sequence 79825–91466; Genbank/EBI Data Bank Accession number CP000039 . 1 ) into the BamHI site of pACYC177 in a direction opposite to that of the tet promoter [26] . The bacterial strains used are listed in S1 Table . Strains MF1632 and MS2834 were constructed as previously described [27] using the primers listed in S2 Table . For immunization or challenge , bacterial cells were grown at 37°C in LB Lenox medium as described previously [27] , concentrated by centrifugation at 3 , 000 × g for 5 min at 4°C , and resuspended in PBS . Immunoblotting was performed as described previously [27] . To examine protein expression at 37°C , three independent cultures of 2457T and MF4835 were harvested at the same growth phase ( OD600 = 0 . 8 ) , blotted onto the same membrane , and levels of IpaB and InvE were measured using a chemical luminescence-based imaging system ( Fusion Solo 7S; VILBER Inc . ) . Values were calculated relative to those of 2457T ( ± the standard deviation ) . HeLa cells were cultured in 6-well plates at 37°C ( 5% CO2 ) in Dulbecco’s modified Eagle’s medium ( DMEM , Invitrogen ) supplemented with 10% fetal bovine serum ( FBS ) . After reaching 60% confluence , Hela cells were washed with sterile PBS and infected with the indicated wild-type and Hfq mutant strains ( MOI = 100 ) . Following infection for 30 min , HeLa cells were washed and treated for 1 h with gentamicin ( 50 μg/ml ) . Finally , cells were permeabilized with 0 . 1% Triton X-100 , serially diluted , and plated on LB plates prior to CFU counts . Six animals were examined in the colon loop model . Three colon segments ( each 4 cm in length ) per animal were tied with a surgical suture . Then , 1 ml of bacterial suspension in PBS ( 1 . 0×109 cfu/segment ) or PBS alone was injected directly into the lumen . Inflammation was observed in each of three animals upon sacrifice at 6 or 24 h post-injection . Tissues were fixed and stained with hematoxylin and eosin . To count the bacteria in the tissues , the intestinal loops were washed twice with PBS containing gentamicin ( 50 μg/ml ) , followed by homogenization in 5 ml of PBS . Serial dilutions ( 10-fold ) were prepared and spread on selective agar plates ( 285310; Hektoen Enteric Agar , Difco ) . Representative colonies were confirmed in an agglutination test for S . flexneri 2a ( 210227; Denka Seiken , Japan ) . The mean number of cfu/g intestinal tissue was calculated from three animals . For bacterial counts in stool suspensions ( 1 g stool/ml PBS ) were calculated as described above . Bacterial strains ( 5 . 0×108 cfu/eye ) were dropped onto the conjunctival sac of both eyes on two consecutive days; this was repeated 14 days later . At 4 days post-first immunization , the development of keratoconjunctivitis was recorded using a digital camera . Animals were then challenged with S . sonnei ( HW383 ) and S . flexneri 1b ( 9268N ) . Results were recorded 4 and 3 days after primary inoculation with HW383 and 9268N , respectively . Animals were euthanized at Day 38 . Ocular tissues were stained with hematoxylin and eosin ( Fig 1E , a ) . Immunization was performed using the same dose ( 5 . 0×108 cfu/eye ) and schedule described in Fig 1E , b , but the left eye was used rather than the right . At Day 28 , animals were challenged with the same dose ( 5 . 0×108 cfu/eye ) of Sd1 ( NT4907 ) , which was placed into the right eye . Keratoconjunctivitis was again observed and recorded at 4 days post-inoculation . On Day 41 , animals were fasted for 24 h . On Day 42 , the ileocecal junction was inoculated with S . sonnei ( IDH00968 ) in PBS ( 1 . 0×109 cfu/ml ) as described previously [33] . Development of watery/bloody diarrhea was monitored for 3 days ( Fig 1E , b ) . Two separate experiments were performed for Sd1 ( NT4907 ) and S . sonnei ( IDH00968 ) . Animals were immunized via the oral route as previously described [34] . Briefly , anesthetized animals were treated with ranitidine ( 50 mg/kg; intramuscularly ) , followed by two 5 ml injections ( 15 min apart ) of 5% sodium bicarbonate directly into the stomach via a sterile human infant feeding tube . MF4835 ( 1 . 0×107 cfu in 1 ml of PBS ) , Wild-type strain 2457T ( 1 . 0×106 cfu in 1 ml of PBS ) , or 1 ml of PBS alone was injected through the same tube and animals were returned to the cage . Body weight , rectal temperature , and bacterial counts ( cfu ) in the stool were measured for 7 days . Immunizations were performed four times , with a 1 week interval between each . At Day 28 , animals were subjected to intestinal challenge with NT4907 or IDH00968 ( 1 . 0×109 cfu ) as described above . After 24 h , symptoms were recorded and three animals were sacrificed to examine intestinal colonization . The body weight , rectal temperature , and survival of the remaining three animals were recorded for 3 days ( Fig 1E , c ) . The levels of IgG and IgA in the serum and sIgA in the stool were measured in ELISA plates coated with strain 2457T , as described previously [33] . Serum/stool samples were serially diluted ( 2-fold ) in PBS ( from 1:100 to 1:12 , 800 ) and a 100 μl aliquot was tested using HRP-conjugated anti-IgG ( A7289; Sigma ) , -IgA ( SA-60-PZ; Immunology Consultant Laboratory Co . , USA ) , or -secretory IgA ( KT-55060; Kamiya Biomedical Co . , USA ) antibodies . TNF-α ( E-90113Gu; USCN Co . , USA ) , IFN-γ ( E-90049Gu; USCN Co . , USA ) , and IL-6 ( MSB701906; MyBioSource Co . , USA ) were measured using ELISA kits . To eliminate non-specific signals , sera were pre-absorbed with formaldehyde-killed S . sonnei ΔinvE strain MS1632 and BL21 bacterial cells , which do not express virulence proteins . Cultures were mixed with formaldehyde ( final concentration: 1% at OD600 = 1 . 0 ) , fixed for 15 min at 37°C , and the cells washed three times with PBS containing 50 mM glycine . Aliquots ( 3 μl ) of pre-immune and immune serum from the three Δhfq-immunized animals were mixed , heated to 56°C for 30 min , diluted ( 250× ) with immunoreaction enhancer solution ( NKB-101; Toyobo , Japan ) containing 0 . 02% NaN3 , and mixed with an excess amounts of killed cells . The pre-absorption ( 37°C for 1 h ) was repeated three times . Absorption by S . sonnei ΔT3SS strain MS2834 and BL21 was also performed . To detect reactive antibodies , bacterial strains were harvested at OD600 = 0 . 4 , collected by centrifugation at 1 , 100×g for 5 min at 4°C , suspended in ice-cold TBS ( 20 mM Tris HCl [pH 7 . 5] , 150 mM NaCl ) , and attached to poly-L-lysine ( P7890; Sigma ) -coated cover glasses for 15 min at 4°C . Cover glasses were fixed in paraformaldehyde ( 4% in PBS ) for 15 min , quenched for 5 min with TBS containing 50 mM glycine , and then blocked with 2% BSA in TBS for 15 min followed by TBS containing 5% normal goat serum ( pretreated at 56°C for 30 min ) for 1 h at 37°C . Cells were washed three times for 5 min with TBS containing 0 . 05% Tween 20 ( TBST ) . Samples were then incubated with pre-absorbed sera for 1 h at 37°C and washed with TBST . AlexaFluora488-conjugated anti-guinea pig IgG ( A11073; Life Technologies ) was diluted ( 100× ) with immunoreaction enhancer solution ( NKB-101 ) and incubated with the samples for 1 h at 37°C . The cover glasses were then washed and sealed with 8 μl of Vectashield ( Vector labs ) and observed under a Carl Zeiss LSM-700 fluorescence microscope ( Carl Zeiss LSM-700 ) . S . sonnei strain HW383 was harvested at OD600 = 0 . 4 , collected by centrifugation at 1 , 100×g for 5 min at 4°C , suspended in ice-cold TBS , and attached to poly-L-lysine-coated glass plates at 4°C . Cover glasses and attached bacteria were placed in 4-well plates and killed in 0 . 4 ml of paraformaldehyde ( 0 . 5% in PBS ) for 15 min , followed by quenching for 5 min in TBS containing 50 mM glycine . Immune serum and serum from unimmunized animals were collected at Day 27 after ocular immunization ( Fig 1E , a ) . Fresh serum was diluted ( 50× ) in TBST , pre-absorbed with the formaldehyde-killed S . sonnei ΔinvE strain MS1632 and BL21 bacterial cells at 4°C for overnight , and incubated with the glass plates for 1 h at 37°C . After washing with PBST , the plates were blocked with 2% BSA and 5% heat-treated goat serum , washed again with PBST , overlaid with a 15 nm gold-conjugated anti-guinea pig antibody ( 815 . 144; Aurion Immuno Gold , 100× dilution in NKB-101 ) , and incubated at 37°C for 1 h . After washing , the plates were fixed with 2 . 5% glutaraldehyde/2% paraformaldehyde and observed under a Hitachi SU6600 scanning electron microscope . Although nearly 5% of cells in samples treated with immunized serum were destroyed , all cells in samples treated with pre-immune serum were intact .
Consistent with our previous studies [27–29] , we found that a mutant of S . flexneri 2a harboring a deletion of the hfq gene ( Δhfq: MF4835 ) expressed InvE and a T3SS effector ( IpaB ) at the repressive temperature of 30°C ( S1A Fig ) . At 37°C , expression was significantly higher than that in the wild-type strain ( Wt: 2457T ) [InvE: × 1 . 67±0 . 29 , p<0 . 05 , IpaB: × 1 . 49±0 . 17 , p<0 . 05 ( n = 3 ) ] , leading to a significantly higher rate of invasion into cultured cell lines ( S1B Fig ) . Also , when 1 . 0×109 cfu of bacteria were injected into colon loop segments in vivo , the mutant caused less severe symptoms ( Fig 1A , a and 1B ) ; also , a high invasion rate was observed at the early stage ( 6 hr ) ( Fig 1D , white bars ) . By contrast , the Wt strain induced bleeding and tissue distraction ( Fig 1A , c and 1C ) . At first , we assumed that effectors of T3SS , such as IpaBCDA , would be the most important antigens; therefore , we transformed the Δhfq mutant with a multi-copy plasmid encoding ipaBCDA genes to increase expression of such antigens ( S1A Fig ) . The strain carrying the ipaBCDA plasmid ( MF4837 ) and the Δhfq and Wt strains were tested by ocular vaccination into Hartley guinea pigs [35] . Bacteria ( 5 . 0×108 cfu / eye ) were dropped into both eyes on two consecutive days; this procedure was repeated 2 weeks later ( Fig 1E , a ) . Animals immunized with Δhfq ( Fig 2A ) or Δhfq harboring the ipaBCDA plasmid ( Fig 2B ) developed keratoconjunctivitis , which was milder than that observed in animals immunized with the Wt strain ( Fig 2C ) . The eyes of all animals recovered by the time of the second immunization , which did not cause a second infection . On Day 28 post-first immunization , animals were challenged in both eyes with the same dose of S . sonnei ( HW383 ) . Unexpectedly , all three animals immunized with the Δhfq strain remained asymptomatic throughout the observation period ( Fig 2E , S2 Fig ) . The five animals immunized with Δhfq carrying the ipaBCDA-encoding plasmid ( Fig 2F ) , as well as all Wt-immunized animals ( Fig 2G ) , developed opaque corneas , although this was more pronounced in the Wt-immunized group . All three animals treated with PBS showed damage to the corneal surface , which was accompanied by excretion of pus ( Fig 2H ) . After recovering from the first challenge with S . sonnei , animals were challenged with S . flexneri 1b strain 9268N ( 5 . 0×108 cfu/eye ) on Day 35 . Two animals in the Δhfq-immunized group were asymptomatic ( S3 Fig ) . The remaining animal showed increased opacity of the left cornea only , but recovered within a few days . Animals immunized with the Wt strain or with Δhfq carrying the ipaBCDA plasmid again developed opaque corneas , although three animals appeared asymptomatic throughout the observation period ( S3 Fig ) . Microscopic examination of the cornea at Day 38 indicated an intact structure in Δhfq-immunized animals ( Fig 2I ) and loss of normal structure in PBS-treated animals ( Fig 2L ) . Animals immunized with the Δhfq strain carrying the ipaBCDA-encoding plasmid ( Fig 2J ) and those immunized with the Wt strain ( Fig 2K ) showed vesicular degeneration and thinning of the eosinophilic layer , both of which were more evident in the Wt-immunized group . The group immunized with bacteria harboring the ipaBCDA-encoding plasmid showed adverse effects , possibly due to changes in optimal T3SS expression due to the multiple ipa genes encoded by the plasmid . Therefore , we used the plasmid-negative Δhfq strain in all subsequent experiments performed at the NICED ( where we used a non-albino breed of guinea pig ) . We next examined whether vaccination protected non-albino guinea pigs against Sd1 . Animals were immunized in the left eye ( 5 . 0×108 cfu ) on Days 0 , 1 , 14 , and 15 , and then subsequently challenged ( on Day 28 ) by administration of the same dose into the right eye ( Fig 1E , b ) . No bacteria were transmitted from the left to the right eye . Animals immunized with the Wt strain ( 2457T ) developed severe keratoconjunctivitis ( Fig 3B; one animal died from systemic infection and three remained infected beyond Day 28 ( S4 Fig ) ) ; however , animals immunized with Δhfq ( MF4835 ) appeared asymptomatic ( Fig 3A ) . Since Hartley guinea pigs developed distinct lesions after immunization with the same Δhfq strain ( Fig 2A ) , it appears that different animal breeds show differing susceptibility to infection . On Day 28 , the right eye of each animal was challenged with the Sd1 strain ( NT4907 ) . The Δhfq-immunized group was again protected against subsequent Sd1 challenge ( Fig 3D ) , which induced keratoconjunctivitis in all PBS-treated animals ( Fig 3F ) . The Wt-immunized group was also asymptomatic ( Fig 3E ) . The finding that Wt-immunized Hartley guinea pigs developed minor symptoms ( Fig 2G ) again highlights potential differences in susceptibility between the two guinea pig breeds . After recovering from the Sd1 infection ( Day 42 ) , animals were subjected to intestinal challenge with S . sonnei strain ( IDH00968 ) via direct inoculation into the colon with concomitant ligation of the cecum to promote infection [33] . PBS-treated animals developed frequent and bloody diarrhea , and all died within 2 days ( Fig 3G ) . By contrast , 2/6 animals in the Δhfq-immunized group and 2/5 in the Wt-immunized group developed mild watery diarrhea , but all recovered within a few days . The remaining animals were asymptomatic . The above results indicate that ocular immunization with Δhfq elicited a systemic immune response against Shigella strains of heterologous serotype . Thus , we next examined an oral immunization model of shigellosis , which was established in achlorhydric animals treated with H2 blockers [34] . Two independent experiments using Sd1 and S . sonnei ( see S1 Text ) were conducted . Animals were immunized with 1 . 0×107 cfu Δhfq ( MF4835 ) /week or with 1 . 0×106 cfu Wt ( 2457T ) /week for 4 weeks ( Fig 1E , c ) . The dose of the Wt strain was reduced to 1 . 0 × 106 cfu due to observed lethality when used at 1 . 0×107 cfu in the preliminary experiment ( see Discussion section ) . At 1 day post-first immunization , the Δhfq-immunized group showed an increase in rectal temperature ( Fig 4A ) [p<0 . 01] and weight loss ( Fig 4B ) [p<0 . 05] without diarrhea; this was not true for the PBS-treated group . The Wt-immunized group developed watery diarrhea , an increased rectal temperature , and weight loss ( maximum 11% at Day 3 when compared with the average for PBS-treated animals ) . Although Δhfq-immunized animal received more bacteria than Wt-immunized animal , fewer were isolated from stool samples on Day 1 [p<0 . 01]; also , the Δhfq-immunized animals cleared bacteria faster than the Wt-immunized group ( Fig 4C ) . After recovering from the first immunization , none of the Δhfq- , Wt- , or PBS-treated animals developed diarrhea after subsequent immunization , suggesting induction of a protective immune response against Shigella strains of the same serotype . Although production of serum IgG ( Fig 4D ) by animals in the Δhfq-immunized group was delayed slightly , the levels at Day 28 were the same as those in the Wt-immunized group . The levels of IgA in the serum ( Fig 4E ) and of secreted IgA ( Fig 4F ) in the stools of the Δhfq-immunized group were similar to those in the Wt-immunized group at Day 28 [p = 0 . 18] . We then compared cytokine levels on Days 7 and 28 . In line with the severity of infection , the levels of TNF-α , IL-6 , and IFN-γ were significantly higher in the Wt-immunized group than in the Δhfq-immunized group ( Fig 4G , 4H , and 4I ) . Immunized animals were subjected to intestinal challenge with Sd1 at Day 28 post-first immunization . All PBS-treated animals excreted frequent watery stools , and 50% developed bloody diarrhea . However , 4/6 animals in both the Δhfq- and Wt-immunized groups were asymptomatic . The remaining animals excreted mucoidal stools , and 50% showed a small amount of bleeding at 24 h post-challenge . At this point , the bacterial counts in intestinal tissues from both immunized groups were significantly lower than those in the PBS controls . However , there was no significant difference in bacterial counts between Δhfq- and Wt-immunized animals ( Fig 4J ) . Data regarding rectal temperature ( Fig 4K ) , body weight ( Fig 4L ) , and survival ( Fig 4M ) supported the generation of a protective immune response in the immunized groups . Microscopic observation of intestinal tissues after Sd1 challenge revealed no evidence of bleeding , although a limited number of erythrocytes was observed in the intestinal lumen of PBS-treated animals . Loss of microvilli and invasion of polymorphonuclear leukocytes into the lamina propria were also observed in PBS-treated animals; tissues from Wt- and Δhfq-immunized animals appeared normal ( Fig 4N ) . The results of the above experiments indicated that the Δhfq strain induced production of protective antibodies against heterologous serotypes of Shigella . Consistent with this , S . sonnei cells were lysed by fresh serum from Δhfq-immunized animals ( Fig 5A ) but not by serum from pre-immune animals ( Fig 5B ) , suggesting complement activation by specific antibodies . Serum samples pre-absorbed with S . sonnei ΔinvE and E . coli BL21 cells reacted with Wt S . sonnei ( Fig 5C ) , Sd1 ( Fig 5D ) , S . flexneri strains of serotype 1b ( Fig 3E ) , 3a ( Fig 5F ) , and 6 ( Fig 5G ) , and with enteroinvasive E . coli ( EIEC ) carrying a virulence plasmid [9] similar to that harbored by Shigella ( Fig 5H ) . Sera did not react with an avirulent Sd1 strain ( Fig 5I ) or with S . flexneri 1b ( Fig 5J ) lacking virulence plasmids , indicating that antibodies are specific for a virulence factor ( s ) encoded by the virulence plasmids . A deletion mutant of S . sonnei ΔT3SS ( MS2834 ΔipaA~spa40 ) was not stained by the sera ( Fig 5K ) , whereas sera pre-absorbed with the ΔT3SS strain and BL21 reacted with Wt S . sonnei ( Fig 5L ) . This indicates that these antibodies at least recognize proteins within the T3SS-encoding region . Finally , S . flexneri 2a ΔinvE strain , which expresses serotype antigens but not virulence proteins ( S1A Fig ) , reacted strongly with serotype-specific antibodies produced upon subsequent immunization with S . flexneri 2a Δhfq ( Fig 5M ) . No signal was detected in Wt S . sonnei reacted with serum from pre-immune animals ( Fig 5N ) .
In the present study , animals immunized with a S . flexneri 2a-based Δhfq strain were protected from heterologous challenge with Sd1 and S . sonnei . The results provide strong evidence supporting cross protection against Shigella strains of heterologous serotypes; these results were replicated in independent guinea pig models . The amount of bacteria ( 5 . 0×108 cfu for ocular immunization and 1 . 0×106–107 cfu for oral immunization ) was much higher than that usually required to cause diarrhea in humans ( 1×102 ~103 cfu ) [36] , leading us to postulate that cross-protection was induced by administration of Shigella strains at excess amounts . Exposure to a sufficient number of bacteria expressing common virulence proteins could induce immunity and broad protection , which may not be fully established during a natural infection cycle during which limited bacteria begin to propagate within intracellular spaces within the colon epithelium , thereby escaping from the host immune system . Indeed , several early studies of a keratoconjunctivitis model documented cross protection , albeit partial , which support this hypothesis . Serény ( who first established this animal model ) and other groups reproducibly documented that keratoconjunctivitis induced partial protection against reinfection of the same eye by Shigella strains of heterologous serotypes , although the other eye was susceptible [37–39] . Also , Adamus et al . report partial and complete induction of systemic immunity in guinea pigs and rabbits , respectively , after subcutaneous immunization with outer membrane proteins from S . sonnei , resulting in protection from subsequent ocular challenge with S . flexneri 3a [40] . These data provide possible clues to understanding the results of studies using strain T32–Istrati [20 , 22] , which harbors three deletions ( ipaBCD , invA [corresponding to the region around spa32] and virG ) in the virulence plasmid [21] . If expression of the T3SS transporter complex ( encoded by mxi-spa region ) itself remains intact , administration of a high dose might support immunization by common virulence proteins . Strain T32-Istrati was later modified by transformation with a virulence plasmid from S . sonnei harboring rfb ( to drive O-antigen biosynthesis ) and two deletions ( ΔvirF and a 37 kbp segment encoding the whole of T3SS {ΔinvE~ipaBCDA~mxi~spa40} ) , leading to production of O-antigens from both S . flexneri 2a and S . sonnei . Field trials showed protection against S . flexneri 2a ( 61 . 07% ) and S . sonnei ( 72 . 48% ) . A review of the literature suggests that it provided 41 . 89% protection against other serotypes [41] , indicating that it is less effective than the original studies suggest . In the context of our hypothesis that cross protection is achieved by immunization with common virulence proteins , the low cross protection efficacy could be attributed to a lack of virulence gene expression after deletion of T3SS and the essential regulator virF , which results in loss of immunogenicity induced by common virulence proteins . A non-human primate ( NHP ) study using 2457T does not support previous data showing cross protection [16] . We have no explanation for this; however , large amounts of bacteria ( 1×1010 cfu ) are generally required to induce onset of diarrhea . Therefore , some researchers consider NHP models unsuitable for evaluation of vaccines [3] . Another possible ( but less likely ) explanation could be differences in the bacterial strains , which were distributed to each laboratory a long time ago . Also , claims that protection was specific to guinea pigs may be based on the fact that successful immunization was achieved by using excess amounts of bacteria . However , these questions require clarification in further studies enrolling human volunteers and using attenuated strains . The colon loop model demonstrated effective attenuation of the Δhfq strain without loss of expression of virulence genes ( Fig 1B ) . Inoculation of the colon loop with excess amounts of bacteria ( 1 . 0×109 cfu ) resulted in a greater number of locally-invading bacteria ( Fig 1D ) without any symptoms , indicating that attenuation afforded by the Δhfq mutation was so effective that fewer side effects emerged , irrespective of the inoculation dose . This is a great advantage in terms of practical use . Experiments using two different breeds of guinea pig highlighted different responses against infection , although the use of two models arose because of difficulties with international transfer of materials . Hartley guinea pigs developed corneal lesions after inoculation with the Δhfq mutant ( Fig 2A ) , whereas non-albino guinea pigs were asymptomatic ( Fig 3A ) . After challenge , Hartley guinea pigs immunized with the Wt strain showed opaque changes in the cornea ( Fig 2G ) , which were not evident in the experiment involving non-albino guinea pigs ( Fig 3E ) . Different breeds of guinea pig show differing susceptibility to infection by pathogens [42] . Also , the difference might be due to the composition of intestinal flora . Hartley guinea pigs were purchased from a commercial farm on which animals were kept under strict sanitary conditions ( including air and diet ) . Non-albino animals were bread at NICED , and so the same level of sanitary control was not possible . Such differences would affect immunity against infection by microorganisms , and possibly by the highly-attenuated Shigella strain . Rodent models of oral infection by Shigella are under development . Here , we developed a new guinea pig model based on achlorhydric treatment; this approach was first used to develop a rabbit model of enterohemorrhagic E . coli infection [34] . Preliminary experiments conducted with 1 . 0×107 cfu of both the Wt and Δhfq strains revealed a high efficacy of infection , resulting in the death of all six Wt-immunized animals . As a challenge that induces lethal damage , an excess amount of bacteria ( 1 . 0×109 cfu ) was directly inoculated into the colon of all animals . Since the body weight of the two immunized groups remained constant , or even increased , at Day 28–30 , the surgical procedure used for the inoculation had a minimal effect on the condition of the animals . Expression of IFN-γ and its receptor increases in patients with shigellosis , and further increases during the convalescence period [43] . The IFN-γ and immunoglobulin responses in animals immunized with the Δhfq strain induced immunity comparable with that induced by the Wt strain . In the two independent experiments , the body weight of all Wt-immunized animals was at least 5% higher than that of the other groups at Day 28 , reflecting the reproducibility of the two experiments . This increase may be due to the stress of severe diarrhea , which might encourage excess uptake of food . Immunological detection of induced antibodies was consistent with the protective effects observed in the challenge studies . Also , detection of antibodies against different strains , including EIEC , suggests positive responses against universal serotypes of Shigella strains . We did not conduct challenge experiments using S . flexneri 2a ( which has a serotype identical to that of the immunizing strain ) because the majority of studies , including one using the S . typhimurium Δhfq vaccine [32] , report generation of immunity against strains of homologous serotype . Production of antibodies against homologous O-antigen was detected using the ΔinvE mutant of S . flexneri 2a strain 2457T ( which has lost expression of all virulence proteins ) , as evidenced by the strongest signal upon immune analysis . Consistent with this , sera from immunized animals strongly agglutinated 2457T cells , but not S . sonnei or Sd1 cells . In addition , Shigella-specific proteins lacking InvE-dependent regulation could be considered potential antigens . However , if these proteins were common among Shigella species , detection of antibodies would be difficult in the experiment that required absorption of serum with the ΔinvE strain of S . sonnei to reduce non-specific signals generated by general bacterial proteins . A rough estimate suggests that vaccines that are effective against both serotypes of S . sonnei and S . flexneri 2a , 3a , 6 will cover about 60% of patients [7 , 44] . Detection of antibodies against these serotypes indicates a potentially broad effect for prevalent strains . In addition , the O-antigen of S . flexneri 2a provides cross protection against S . flexneri strains 1a , 2b , 3b , 4a , 5a , and Y , all of which possess group factor 3 , 4 and type factor II [45] . This provides a S . flexneri 2a-based vaccine with a great advantage over other vaccine candidates that target a limited number of virulence proteins . Immunization with common virulence proteins from an attenuated mutant is a new concept; fortunately , a single mutation attenuated the immunizing strain and increased antigen expression . These results suggest that such a strategy could be applied to other pathogens harboring common virulence machinery if one carefully selects the appropriate strain and mutation to provide effective attenuation without loss of antigen expression . Common virulence protein antigens expressed by the attenuated strain and acting as “live toxoids” are expected to elicit the same levels of host immunity against multiple serotypes of pathogen as that elicited by conventional toxoids . | An ideal vaccine should show a broad range of protective efficacy against all serotypes of a particular pathogen . Variations in the structural lipopolysaccharide “O”-antigen mean that it is difficult to induce protection against multiple serotype Shigella strains using a single serotype immunogen . Here , we examined vaccination effects of an attenuated S . flexneri hfq mutant strain overexpressing virulence proteins common to all Shigella serotypes , which was identified from our studies on temperature and osmotic-dependent regulation of the type III secretion system: an essential virulence machinery among Shigella and enteroinvasive Escherichia coli species . The deletion mutant lacked the RNA-binding protein Hfq , which led to increased expression of the type III secretion system via loss of regulation , resulting in attenuation of cell viability through repression of stress response sigma factors . Immunization with the hfq mutant of S . flexneri 2a induced systemic immunity and cross-protective efficacy against two different serotypes of Shigella strain: S . dysenteriae type 1 and . S . sonnei . Detection of anti-Shigella antibodies against various strains with different serotypes suggests that generation of antibodies specific for common virulence proteins affords cross-protection against Shigella strains of multiple serotypes . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"tropical",
"diseases",
"microbiology",
"vertebrates",
"animals",
"mammals",
"shigella",
"animal",
"models",
"vaccines",
"bacterial",
... | 2017 | An attenuated Shigella mutant lacking the RNA-binding protein Hfq provides cross-protection against Shigella strains of broad serotype |
Canine leishmaniasis ( CanL ) is a zoonotic disease , caused by Leishmania infantum and transmitted by Phlebotomus perniciosus in the Mediterranean basin . Previously , an ELISA based on the P . perniciosus salivary protein SP03B was proposed as a valid tool to screen for canine exposure to sand fly bites across regions endemic for CanL . Although this approach is useful in laboratory settings , a practical tool for immediate application in the field is needed . In this study we propose the rSP03B sero-strip , the first immunochromatographic test ( ICT ) in the field of vector exposure able to rapidly screen dogs living in endemic areas for the presence of P . perniciosus and to aid in the evaluation of vector control programs . The ICT was prepared using the bacterially expressed recombinant protein rSP03B as antigen . For test optimization , pre-immune sera from non-bitten laboratory-bred Beagles were used as negative controls . In order to validate the test , sera from laboratory-bred Beagles experimentally exposed to P . perniciosus bites were used as positive controls . Additionally , all samples were tested by ELISA using whole salivary gland homogenate ( SGH ) and the rSP03B protein as antigen . An almost perfect degree of agreement was found between the ICT and the SGH-ELISA . Furthermore , the newly proposed rSP03B sero-strip showed a sensitivity of 100% and a specificity of 86 . 79% . We developed a simple and rapid ICT based on the P . perniciosus rSP03B salivary protein , able to replace the standard ELISA used in previous studies . Our rSP03B sero-strip showed to be highly sensitive and specific in the detection of antibodies ( IgG ) against P . perniciosus saliva . In the future , this test can be employed during large-scale epidemiological studies of CanL in the Mediterranean area to evaluate the efficacy of vector control programs .
Canine leishmaniasis ( CanL ) is a widespread zoonotic disease present in several countries in Latin-America , Europe and Asia [1 , 2] . It is a severe multi-systemic disease of dogs caused by the protozoan parasite Leishmania infantum . The disease manifests itself in variable clinical signs , with the majority of dogs experiencing poor body condition , generalized muscular atrophy , lymphadenomegaly and excessive skin scaling ( reviewed in [3] ) . CanL is endemic across the Mediterranean basin [1 , 2] , with seroprevalences varying from region to region depending on ecological aspects [4] . Overall , 2 . 5 million dogs are estimated to be infected in southern Europe [4] . Since dogs suffering from the disease are extremely difficult to treat , it is not surprising that the high incidence of CanL in southern Europe represents the main cause of deaths amongst dogs in the region [5] . However , recent occurrence of autochtonous cases in Romania , Hungary and northern Italy suggests that the disease is not limited anymore to the Mediterranean region , but confirms its spread to more northern areas ( reviewed in [6] ) . It is noteworthy that less than 50% of infected dogs develop the disease [7] . However , both sick and asymptomatic dogs represent the main reservoir of the parasite and form a risk for human disease , zoonotic visceral leishmaniasis ( ZVL ) [3 , 5] . In the Mediterranean region , the annual incidence of ZVL is estimated to range from 1 , 200 up to 2 , 000 [8] . CanL endemicity is associated with the distribution and abundancy of its vectors , phlebotomine sand flies . In southern Europe , 5 species are proven vectors of CanL [9] , of which Phlebotomus perniciosus is the most important . During the bite , the sand fly injects saliva containing a cocktail of bio-active molecules with anti-hemostatic , anti-inflammatory and immune-modulatory activities into the host skin ( reviewed in [10] ) . These molecules facilitate the blood-feeding process of the sand fly and trigger a humoral immune response in the host . It is well-known that the amount of host anti-saliva IgG antibodies ( Abs ) correlates with the level of exposure to sand flies [11 , 12] . Furthermore , previous studies showed a clear fluctuation of the Ab response during longitudinal sampling of dogs over two transmission seasons [26] , suggesting that proteins present in sand fly saliva can be a useful tool to evaluate the efficacy of vector control programs . For example , previous studies on mosquitoes [13–15] and triatomine bugs [16] have shown that a reduction in vector density observed after the implementation of insecticide treated nets ( ITNs ) correlates with a reduction in anti-vector salivary Ab-response . With regard to sand flies , only one study performed in India and Nepal measured anti-P . argentipes Abs to evaluate the use of ITNs [17] . Performing large-scale serological studies to detect host exposure to sand fly bites was limited in the past due to the fact that dissecting large amounts of sand fly salivary glands is a demanding and labour-intensive process . Besides , the use of whole SGH is subject to protein content variability , dependent on the age of the sand fly at the time of dissection [18] and might antigenically cross-react with taxonomically closely related sand fly species ( reviewed in [19] ) . Therefore , using specific antigenic recombinant sand fly salivary proteins as a replacement to the use of whole SGH has gained more attention [20–23] . Previously the specific antibody response ( IgG ) against the salivary protein SP03B from P . perniciosus was proposed as a valid exposure marker across regions endemic for CanL [22] . This study demonstrated the presence of similar antigenic epitopes in the recombinant SP03B protein compared to its native form , and indicated a substantial antigenic cross-reactivity amongst P . perniciosus populations from Campania , Umbria and the metropolitan Lisbon region [22] . The SP03B salivary protein belongs to the family of yellow-related proteins [24] and was previously shown to possess binding activity for pro-haemostatic and pro-inflammatory biogenic amines in Lutzomyia longipalpis [25] . Recently , recombinant yellow-related proteins were subject of epidemiological studies to determine the levels of specific anti-vector salivary Abs in naturally bitten hosts [20 , 21 , 26 , 27] . All of these studies used indirect enzyme-linked immunosorbent assays ( ELISA ) . Although this approach is useful in laboratory settings , a practical tool that can immediately be used in the field and that allows a fast screening of hosts living in endemic areas is called for . In this study , we evaluated our newly proposed colloidal gold immunochromatographic test ( ICT ) –the rSP03B sero-strip—against the standard ELISA method . The use of colloidal gold ICTs has first been described by Osikowicz and Beggs for the qualitative detection of human chorionic gonadotropin ( hCG ) [28] and since then deployed in a broad range of fields . However , with regard to the detection of IgG Abs against arthropod saliva , no such test has been described yet . Therefore , we propose the first colloidal gold ICT in the field of vector exposure . Since our test is very straightforward—no special equipment or skill is required—it can be easily operated in non-laboratory settings to rapidly screen large cohorts of dogs for exposure to P . perniciosus .
Reuse of canine sera samples obtained during previous studies [11 , 26] was approved by the Ethical Board of Charles University . Ethical approval of sera samples collected from non-bitten laboratory-bred Beagles housed at the University of Zaragoza , Spain ( UNIZAR ) was obtained during another ongoing study , protocol PI44/17 . All sampling complied with the European guidelines on the protection of animals ( Directive 2010/ 63/UE ) . Forty-two sera samples from laboratory-bred Beagles experimentally exposed to P . perniciosus were used to prepare the first prototype of the test . These dogs were individually exposed to approximately 200 P . perniciosus females during a previous study . The sampling protocol is described in more detail in [11] . Negative control sera were collected from 29 non-bitten laboratory-bred Beagles housed at the University of Zaragoza , Spain ( UNIZAR ) . Furthermore , sera samples from 24 laboratory-bred Beagles born in a breeding facility located in northern France were used as negative controls . The antigen ( Ag ) used for the preparation of the rSP03B sero-strip is a bacterially expressed 43kDa yellow-related recombinant protein of P . perniciosus ( rSP03B , Genbank accn . DQ150622 ) . The recombinant protein was obtained from Apronex s . r . o . ( Prague ) as described in [29] and was expressed with the Escherichia coli BL21 ( DE3 ) expression system in the pET28b vector ( Novagen ) with a poly-His tag ( 6 histidines ) . The protein was isolated under denaturing conditions with 8M urea ( 50mM Tris , pH 8 , 300mM sodium chloride ) and prepared for usage in the rSP03B sero-strip by gradually dialyzing it to a final concentration of 0M urea ( PBS 1x , pH 6 ) using Slide-A-Lyzer mini dialysis units ( 10K MWCO , 0 . 1mL ) , following the manufacturer’s protocol . The UV absorbance value of the protein was determined by Nanodrop at 280nm . The protein concentration was then quantified by means of the known molar extinction coefficient of the protein . The Ags used for the ELISA include the rSP03B salivary protein and the whole salivary gland homogenate ( SGH ) from P . perniciosus . A colony of P . perniciosus was reared under standard conditions as described in [30] and salivary glands were dissected from 4–6 days-old female sand flies , pooled in 20mM Tris buffer with 150mM NaCl and stored at -20°C . Before use , the SGH was prepared by disrupting the salivary glands during 3 freeze-and-thaw cycles in liquid nitrogen . All sera samples were analyzed by an indirect enzyme-linked immunosorbent assay ( ELISA ) that measures anti-P . perniciosus IgG . The ELISA was performed in accordance with previous studies [26] , with minor modifications . Briefly , flat bottom microtiter plates ( Immulon ) were coated with P . perniciosus salivary gland homogenate ( SGH ) ( 0 . 2 salivary gland per well ) or with rSP03B ( 5μg/ml ) in 20mM carbonate-bicarbonate buffer ( pH 9 , 100μl/well ) and incubated overnight at 4°C . The plates were washed with PBS + 0 . 05% Tween 20 ( PBS-Tw ) and blocked with 6% ( w/v ) low fat dry milk diluted in PBS-Tw . Canine sera diluted in 2% ( w/v ) low fat dry milk/PBS-Tw was added to the wells ( 100μl/well ) . Sera were diluted at 1/200 and 1/100 for SGH and rSP03B , respectively . After 90min incubation at 37°C , the plates were incubated at 37°C for 45min with secondary Abs ( polyclonal anti-dog IgG-horseradish peroxidase ( HRP ) , Bethyl laboratories , 100μl/well ) diluted 1:9000 in PBS-Tw . The ELISA was developed using an orthophenylendiamine ( OPD ) solution in a phosphate-citrate buffer ( pH 5 . 5 ) with 0 . 1% hydrogen peroxide . The reaction was stopped after 5min with 10% sulfuric acid and absorbance ( OD value ) was measured at 492nm using a Tecan Infinite M200 microplate reader ( Schoeller ) . Each serum was tested in duplicate . The rSP03B sero-strip is composed of a lower absorbent pad and an upper absorbent pad that both overlap a nitrocellulose ( NC ) membrane located in the middle of the test ( Fig 1 ) . The lower absorbent pad is impregnated with a colloidal gold-conjugate consisting of a mixture of one conjugate for the test line and one for the control line . The conjugate for the test line was prepared by coupling a polyclonal anti-dog IgG Ab ( Bethyl laboratories ) to colloidal gold nanoparticles . Secondly , the control conjugate was prepared by coupling a chicken Ab from non-immunized chickens to colloidal gold nanoparticles . The coupling of both conjugates was followed by a saturation step ( gold blocking buffer , Coris BioConcept ) . On the NC membrane 3 lines were coated . The first line consists of sample deposition line and enables a complete migration of the sample . The second line is the test line on which the dialyzed rSP03B protein is coated ( 0 . 6mg/ml , 0 . 1μl/mm ) and the third line ( migration control ) consists of a goat anti-chicken Ab ( GAC ) that binds to the colloidal gold ‘control’ conjugate . In order to launch the test , a buffer ( 3μl; HC dilution buffer , Coris BioConcept ) is applied to the sample line ( blue ) on the NC membrane ( step 1 in Fig 2 ) . Immediately after applying the buffer , the serum sample ( 1μl ) is deposited on the same spot ( step 2 in Fig 2 ) . This will cause the sample to start migrating to the upper part of the strip , where the anti-rSP03B Abs present in the serum sample will be captured by the rSP03B Ag coated on the test line ( step 3 and 4 in Fig 2 ) . Directly after deposition of the sample , the strip is dipped into the migration buffer solution ( step 4 in Fig 2; Ly-B dilution buffer , Coris BioConcept ) . The colloidal gold conjugate , consisting of the test- and control-conjugate mixture , gets hydrated and starts migrating upwards together with the moving liquid ( step 5 in Fig 2 ) . Once arrived at the test line , the conjugate will recognize dog IgG Abs bound to the rSP03B Ag coated at the test line ( step 6 in Fig 2 ) . While moving further up the NC membrane , the GAC Abs present at the control line will capture the colloidal gold ‘control’ conjugate , resulting in the appearance of a purple color . This is an essential part of the rSP03B sero-strip as it ensures that the migration went well and the strip is functioning properly . The test is run for 15min and excess buffer solution is absorbed by the upper absorption pad . Intensity of the purple color at the test line relates to the amount of target Ab present in the sample and is visually inspected ( Fig 3 ) . The test is only valid if the migration control line is visible . A positive test result is observed when two lines ( test and control ) are visible on the NC membrane . The test is considered negative when only the control line is present ( Fig 4 ) . All samples tested in ELISA were tested in duplicate . The coefficient of variation ( CoV ) was calculated for each duplicate sample . If the CoV was more than 15% the sample was retested . Since plates were ran on multiple days , positive control sera ( PC , n = 2 ) and negative control sera ( NC , n = 2 ) were included in each plate . In order to correct for interplate variability , the obtained OD values were standardized ( SOD ) according to the following formula: SOD ( % ) =ODsampleAverageODPCx100 ELISA cut-offs were determined by the mean of the OD values plus 3 standard deviations ( SD ) obtained from non-bitten negative controls . Graphical representation of the distribution of the SOD ( % ) was visualized using the “beeswarm” package in R software [31 , 32] . The results of the rSP03B sero-strip were classified according to the intensity of the observed band , starting from ( 0 ) a negative test result; ( 1 ) a very faint signal; ( 2 ) a low positive signal; ( 3 ) a positive but less intense signal than the control band and ( 4 ) a strong positive signal , same intensity as the control band . All samples classified in categories ( 1 ) , ( 2 ) , ( 3 ) and ( 4 ) were considered positive . The data was graphically represented using the “ggplot2” package in R software [31 , 33] . The degree of agreement between SGH-ELISA , rSP03B-ELISA and the rSP03B sero-strip was measured by Cohen’s Kappa according to the methods of Jacob Cohen [34] . Furthermore , the percentage of agreement between different serological methods was measured and the Mc Nemar’s χ2 test in R software [31] was used to test for significant differences in agreement between the golden standard SGH-ELISA , the rSP03B-ELISA and the rSP03B sero-strip . Correlation between both ELISA tests was analyzed using Pearson’s r correlation coefficient and graphically visualized using the “ggplot2” package in R software [31 , 33] .
The specificity and sensitivity of the 3 serological methods used were calculated based on the results from experimentally exposed dogs and negative control sera ( Table 1 ) . The distribution of the results in SOD ( % ) is shown in Fig 5 . Results from the SGH-ELISA and rSP03B-ELISA were first classified as being positive or negative according to their respective cut-off value ( Fig 5A and 5B ) . All 42 sera of dogs experimentally exposed to P . perniciosus appeared positive on the SGH-ELISA . However , only 29 out of these 42 samples were classified as being positive in the rSP03B-ELISA . Additionally , a single serum sample out of 53 negative control sera was found to be positive in the SGH-ELISA , and rSP03B-ELISA . These observations resulted in a sensitivity of 100% and 69% for the SGH-ELISA and rSP03B-ELISA , respectively , and a specificity of 98 . 11% for both methods . Results obtained from our newly prepared rSP03B sero-strip were classified according to the intensity of the observed band . The distribution of the samples tested by the rSP03B sero-strip is shown in Fig 5C . All 42 sera samples of experimentally exposed dogs presented 2 purple bands—one on the test line and one on the control line—resulting in a sensitivity of 100% . On the other hand , 7 out of 53 sera from non-bitten negative controls were also found positive with our rSP03B sero-strip , giving it a specificity of 86 . 79% . Importantly , raising the detection limit of the rSP03B sero-strip until category ( 1 ) will be considered negative results in a specificity of 96 . 23% , without changing the sensitivity . Since the SGH-ELISA showed the highest sensitivity and specificity amongst all 3 methods , it was set as the golden standard against which the rSP03B-ELISA and the rSP03B sero-strip were evaluated ( Table 2 ) . The sensitivity and specificity of the rSP03B sero-strip as compared to the SGH-ELISA were 97 . 67% and 87% , respectively . A percentage of agreement of 91 . 58% was obtained with no systematic difference between the proportions of positive responses from these two methods ( McNemar χ2 P > 0 . 05 ) . Furthermore , the Cohen’s kappa value between the rSP03B sero-strip and the SGH-ELISA was 0 . 83 , suggesting an almost perfect strength of agreement between these two methods . With regard to the rSP03B-ELISA a sensitivity and specificity of 67 . 44% and 98% were measured , respectively . The percentage of agreement between the rSP03B-ELISA and the SGH-ELISA was set at 84 . 21% with a Cohen’s kappa value of 0 . 67 , suggesting a substantial strength of agreement . However , McNemar χ2 tested a significant systematic difference between the proportions of positive responses from the rSP03B-ELISA and the SGH-ELISA ( P < 0 . 05 ) . Interestingly , when raw OD-values of both ELISA methods were compared , a correlation of 83 . 43% was obtained ( 95% CI [76 . 06%– 88 . 67%] , P < 0 . 001 ) ( Fig 5D ) .
Previous studies addressing the level of host exposure to sand fly bites consistently used ELISA methods to determine the levels of specific anti-vector salivary Abs . Although this method is useful in laboratory settings , a rapid test that can aid in vector control by allowing a consistent screening of hosts in the field has not yet been described . Here , we developed a new rapid test that can be immediately used in the field to screen dogs living in endemic CanL areas for the presence of anti-P . perniciosus IgG Abs . In the proposed rSP03B sero-strip , the yellow-related rSP03B P . perniciosus salivary protein was used as Ag , previously proposed as a valid exposure marker for P . perniciosus across its entire area of distribution [22] . The principle of the rSP03B sero-strip is similar to an indirect ELISA; first specific canine Abs present in the sample bind to the P . perniciosus salivary Ag immobilized on the NC membrane of the sero-strip , after which the interaction is visualized by an anti-dog IgG Ab gold-conjugate . In this study , 3 serological methods to define the level of P . perniciosus salivary IgG Abs were compared: the golden standard SGH-ELISA , the previously described rSP03B-ELISA [22] and our newly proposed rSP03B sero-strip . The results highlight the SGH-ELISA as being the most sensitive ( 100% ) and most specific ( 98 . 11% ) amongst all 3 methods . When comparing the performance values of the rSP03B-ELISA and the rSP03B sero-strip , the rSP03B sero-strip appears to have the highest sensitivity ( 69% vs . 100% , respectively ) whereas the rSP03B-ELISA shows the highest specificity ( 98 . 11% vs . 86 . 79% , respectively ) . Finally , the rSP03B sero-strip was shown to have an almost perfect agreement with the SGH-ELISA without any significant differences , overall suggesting that the sero-strip performs better than the rSP03B-ELISA . Moreover , the performance of the sero-strip can be further improved as was shown by an increased specificity ( up to 96% ) without a loss in sensitivity when very faint results would be considered negative . This can be achieved by the use of a strip reader which will classify the results according to a pre-defined cut-off value for the sero-strip . Although our results indicate that our newly proposed rSP03B sero-strip is a valid replacement for the rSP03B-ELISA , the performance values of the rSP03B-ELISA should be taken with caution . When raw OD-values of both ELISA methods were compared , a correlation of 83 . 43% was achieved . This high correlation between the rSP03B-ELISA and the SGH-ELISA suggests that classifying the samples according to positivity might be the reason for the high number of false negatives in the rSP03B-ELISA . The higher cut-off value observed in the rSP03B-ELISA could be explained by a non-specific reaction that takes place in the ELISA between the negative control samples and bacterial proteins that possibly co-purified with the rSP03B protein [35] and might be overcome by producing the protein in a different expression system ( e . g . mammalian cells ) . The reason why these false negative results do not occur to the same extent with the rSP03B sero-strip is explained by the fact that this is a qualitative method and is therefore not dependent on a specific cut-off value . Thus , any observed signal will be classified as positive . In summary , we developed a simple and rapid colloidal gold ICT based on the bacterially expressed recombinant protein rSP03B that is able to replace the ELISA method used in numerous previous studies [11 , 20–22 , 26] . Our rSP03B sero-strip showed to be highly sensitive ( 100% ) and specific ( 86 . 79% ) in the detection of IgG Abs against P . perniciosus saliva . The test is easily operated with no requirements for skilled personnel or specialized equipment . However , in order to confirm the field detection accuracy and applicability of the test , further evaluation of canine populations exposed to various frequencies of sand fly bites and validation of the test with whole canine blood is required . Additionally , it is worth to mention that potential cross-reactivities between Abs recognizing salivary proteins of closely related sand fly species has been observed in previous studies [36] . Therefore , a cross-reaction between the rSP03B protein and Abs against salivary proteins of closely related sand fly species of subgenus Larrousius is likely to occur , hence rendering the sero-strip useful for estimating exposure to other vectors of CanL in the Mediterranean basin . However , due to the lack of colonies of sand fly species co-occurring with P . perniciosus [37] this cross-reaction cannot be tested at the moment . Unfavorable cross-reactions with other haematophagous insects are very unlikely but we suggest to reflect on this during further studies . | The sand fly Phlebotomus perniciosus is the principle vector of Leishmania infantum , causing canine leishmaniasis in the Mediterranean basin . While the sand fly female takes a blood meal , it injects saliva into the host skin , evoking a specific antibody response in the host . The antibody level in the host correlates with the intensity of exposure to sand flies . Previously , the specific antibody response ( IgG ) against a salivary protein of P . perniciosus—SP03B—has been proposed as a valid biomarker to estimate dog exposure to P . perniciosus in the Mediterranean area . Since standard serological methods are impractical and time-consuming in field conditions , we propose the rSP03B sero-strip—a rapid test that can be immediately applied to screen large cohorts of dogs for the presence of anti-P . perniciosus antibodies . Our test is the first rapid test in the field of vector exposure , it is highly sensitive and specific and shown to be a valid replacement for standard serological assays . In addition , this test could be used as an evaluation tool for vector control programs . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"vertebrates",
"sand",
"flies",
"saliva",
"dogs",
"animals",
"mammals",
"infectious",
"disease",
"control",
"insect",
"vectors",
"... | 2018 | Evaluation of the rSP03B sero-strip, a newly proposed rapid test for canine exposure to Phlebotomus perniciosus, vector of Leishmania infantum |
Trypanosomatid parasites cause serious infections in humans and production losses in livestock . Due to the high divergence from other eukaryotes , such as humans and model organisms , the functional roles of many trypanosomatid proteins cannot be predicted by homology-based methods , rendering a significant portion of their proteins as uncharacterized . Recent technological advances have led to the availability of multiple systematic and genome-wide datasets on trypanosomatid parasites that are informative regarding the biological role ( s ) of their proteins . Here , we report TrypsNetDB ( http://trypsNetDB . org ) , a web-based resource for the functional annotation of 16 different species/strains of trypanosomatid parasites . The database not only visualizes the network context of the queried protein ( s ) in an intuitive way but also examines the response of the represented network in more than 50 different biological contexts and its enrichment for various biological terms and pathways , protein sequence signatures , and potential RNA regulatory elements . The interactome core of the database , as of Jan 23 , 2017 , contains 101 , 187 interactions among 13 , 395 trypanosomatid proteins inferred from 97 genome-wide and focused studies on the interactome of these organisms .
Trypanosomatid parasites cause life-threatening diseases in humans and major production losses in animals . They pose global threats , and various issues are associated with available drugs against trypanosomatids ( including tolerability , cost , and resistance ) , necessitating the identification of novel essential parasitic-specific pathways/genes as potential drug targets [1] . However , as supported by whole genome sequencing data , it is well known that species of the trypanosomatid family , while showing high similarity in proteomes with one another , are highly diverged from other eukaryotes [2–5] . This makes the annotation transfer of nearly half of their proteome by homology-based approaches from model organisms unreliable [3] . During the past decade , several genome-wide and focused studies have been conducted to functionally characterize trypanosomatid proteins . The construction of global and local protein interaction maps has served as one of the main resources for functional annotation by reflecting the molecular context of proteins in a cell [6–15] . Several experimental techniques exist to identify the interacting partners of proteins that differ in selectivity and sensitivity . Therefore , one major challenge in the study of protein interactions is the ability to distinguish between the correctly associated proteins from confounding elements that are present in the results of these experiments . It is also helpful to know the potential interacting proteins that are missing from the results of an experiment based on previously known knowledge of the species of interest or other related trypanosomatid species . Several databases have been developed to represent the experimentally identified or computationally inferred physical and functional protein interactions [16–21] . Such databases greatly help researchers to interrogate cellular processes and gain a systems level view of the protein ( s ) of choice . Although it is of critical importance for studies on trypanosomatids , only a limited number of databases cover information on protein interactions of these parasites , and such interactions are mostly predicted by transferring the available data from other eukaryotes , missing most parts of the published data on trypanosomatid species [16] . Another major approach for the functional characterization of proteins stems from recent technological advances that have allowed measuring transcriptome , proteome , and transcript half-life changes in response to environmental changes , different life stages , or cell conditions [8 , 22–38] . Moreover , it is possible to gain insights on the function of a protein by gathering information on: 1 ) its annotation from resources such as gene ontology , KEGG pathways , and the BioCyc database [37 , 39]; 2 ) protein characteristics , such as protein sequence motifs , isoelectric points , molecular weights , and the number of transmembrane domains; 3 ) the essentiality of gene knock-down on cell survival [33]; and 4 ) the potential cis-regulatory elements present in the 3′-UTR of the gene and the collective response of genes containing that regulatory element to environmental changes [29] . Currently , the TriTrypDB database is a gene-centric framework devoted to the kinetoplastid parasites and provides extensive information on the queried protein ranging from genomic sequence and position to involved biological pathways and captured responses in previously reported studies [38] . However , in many cases , researchers are interested in knowing the collective response of a list of pre-specified proteins along with their interacting partners according to large-scale studies rather than focusing on one protein . Combining interaction data with enrichment analyses of gene ontology , molecular pathways , gene essentiality , and protein sequence features is the key to perceiving the function of proteins . Here , we describe TrypsNetDB , a user friendly , integrated database that fills the aforementioned gaps by not only depositing the current interactome knowledge on trypanosomatid proteins but also combining such information with other available resources accompanied with related statistical analyses . Moreover , the database automatically performs inter-species mapping of the available data and provides information to allow for a better characterization of the queried proteins in the species of interest . Finally , based on the built-in features , the database can help researchers with their interactome related experiments by distinguishing the likely binding partners of a protein from confounding elements identified in their experiments and suggesting other potentially interacting proteins that are missing from the list of queried proteins . Built on powerful ASP . Net framework , the database performance is fast and reliable . TrypsNetDB is freely available at trypsNetDB . org .
The current release of the database is focused on physical protein interaction data that are already published in the trypanosomatid field , supporting 16 trypanosomatid parasites including T . cruzi strain CL Brener , T . cruzi CL Brener Esmeraldo-like , T . cruzi CL Brener Non-Esmeraldo-like , T . brucei gambiense DAL972 , T . brucei Lister strain 427 , T . vivax Y486 , T . evansi strain STIB 805 , T . brucei TREU927 , L . major strain Friedlin , L . mexicana MHOM/GT/2001/U1103 , L . infantum JPCM5 , L . donovani BPK282A1 , L . braziliensis MHOM/BR/75/M2903 , L . braziliensis MHOM/BR/75/M2904 , L . arabica strain LEM1108 , and L . enriettii strain LEM3045 . Fig 1 represents the schematic architecture of the database . To systematically extract the protein interaction data , we searched the NCBI PubMed database using the keywords Trypanosoma and Leishmania and extracted all resultant abstracts . Next , by a manual search , initial positive and negative gold standard sets were constructed by considering 46 and 251 articles , respectively . A multinomial naïve Bayes classifier was used to prioritize 6581 articles that were more likely to contain protein interaction data based on the abstract content with an estimated probability greater than 0 . 75 . By a manual inspection of some articles , the initial positive and negative gold standard set was expanded to 60 and 332 articles , respectively ( with extra attention on keeping the diversity of the gold standard sets to reduce the chance of biased predictions ) . The multinomial naïve Bayes classifier was re-trained and then re-applied to all the extracted abstracts from PubMed using the new gold standard set . A total of 1996 articles that were likely to include interaction data were identified ( estimated probability of 0 . 9 ) . By reviewing the articles of the final list , we could extract protein interaction data from 97 different studies . The interaction data were obtained using a variety of techniques , including affinity purification , immunoprecipitation , yeast two-hybrid ( Y2H ) , fractionation patterns , and other possible experimental techniques . We have only considered the syntenic orthologs reported by TriTrypDB to transfer the inter-species information . Users can query the database based on either tritrypDB IDs ( recognizing IDs of recent and older versions of the database ) or gene names . Support for the remainder of the trypanosomatid species is scheduled to be added in the coming months . In cases where the gene names match multiple organisms , the user will be asked to select the species of interest from a dropdown box . As shown in Fig 2 , querying of protein ( s ) will redirect the user to the interaction page , which is composed of the following three main elements: information panel , network , and reference section . The information panel can be used to explore the details of the three sections of annotations , protein descriptions , and the constructed network . The annotation tab contains gene set enrichment analyses of the combined set of queried proteins with the suggested proteins by the database ( i . e . , proteins with gray background ) for enrichment in the following five different categories: 1 ) GO & Pathway: genes are examined for enrichment in Gene Ontology , KEGG pathway , and BioCyc annotations using hypergeometric tests . Terms with Benjamini-Hochberg corrected p-values less than 0 . 05 are reported back to the user . Hovering over each term will highlight the proteins that are associated with the term . Clicking on each represented term will show a description of the term , its category , number of proteins in the network associated with that term , and the corresponding corrected p-value . 2 ) Sequence & Structure: The sequence and structural features of the proteins are examined , such as the protein motifs , isoelectric points , molecular weights , and predicted number of transmembrane domains ( statistical test for protein motifs is based on the hypergeometric test , and for the other categories based on Wilcoxon-Mann-Whitney rank sum test ) . This information can provide a complementary view of the function of the proteins . For example , a group of soluble interacting proteins are expected to have a significantly low number of transmembrane domains . Likewise , proteins interacting with RNA and DNA are expected to have high isoelectric points . Similar to the GO and pathway enrichment results , hovering over significantly enriched protein motifs will highlight the associated proteins in the network . 3 ) Expression patterns: the proteins in the network are examined for their collective transcriptome and proteome responses across 48 distinct samples using Wilcoxon-Mann-Whitney rank sum test . Each sample is color coded with yellow and blue indicating over-expression/enrichment and under-expression/depletion , respectively . Statistically significant terms with p-values less than 0 . 05 are highlighted by darker colors , while non-significant conditions are semi-transparent . The 48 considered cell states were obtained from genome-wide experiments on T . brucei , T . cruzi , and L . infantum [22–24 , 26–28 , 30–32 , 34–36 , 40 , 41] . By considering syntenic orthologs ( as defined in TriTrypDB ) , the database automatically propagates the information to other trypanosomatid species . Clicking on each sample will open information on the title of the sample , a description of the results of the statistical tests , the calculated p-value , the title of the study that published the sample and its PMID with a link to the PubMed abstracts . 4 ) Gene essentiality: The essentiality of proteins in four different cell conditions of T . brucei are examined based by application of hypergeometric tests on the results of a genome-wide phenotyping study [33] . Ortholog mapping is performed for cases in which the queried organism is not T . brucei . 5 ) 3′ regulatory elements: Using a novel approach , we recently predicted 88 cis-regulatory elements that are potentially involved in the developmental regulation of T . brucei [29] . Although only a limited number of functional elements have been identified thus far , by a rigorous analysis of results , we showed that 11 predicted motifs strikingly resemble previously identified regulatory elements in trypanosomatids , suggesting the high accuracy of the predictions . This section examines whether the 3′-UTRs of the orthologs in the set of proteins in T . brucei are significantly enriched for any of the predicted 88 motifs using hypergeometric tests . In cases where enrichment is found , the motif logo along with the transcriptome and proteome responses of the motif in different cell conditions are reported . The proteins tab provides brief information ( such as transcript and protein length , isoelectric point , molecular weight , etc . ) with a link to the TriTrypDB database in a sorted way , starting from the queried proteins and ending with the proteins that were included in the network by the program ( these suggested proteins have been highly connected to the queried proteins based on literature derived interactions ) . The network tab can be used to explore the contribution of each experimental technique to the construction of the illustrated network . In two cases of tagged affinity purification and immunoprecipitation in which interactions can show indirect associations , the database distinguishes between interactions that are identified based on RNase treatment of the samples from those that are not . Hovering over each technique will highlight the interactions that they support . It is also possible to filter some of the techniques by unchecking the corresponding checkboxes and clicking on the “set filters” button . The network section , using a dynamically interactive interface , represents the interactions among the proteins with each protein indicated by a circular node . It is possible to zoom in or out of the network and reposition the proteins . Queried proteins and other proteins suggested by the database are shown in blue and gray , respectively . The node size of proteins indicates the number of interactions that they have in the global network with larger nodes representing nodes with a higher number of interactions . Selecting a protein by clicking on it will highlight the first neighbors of that protein and open the corresponding information in the proteins tab of the information panel . Finally , the network option on the top-left part of the network section can be used to automatically rearrange the network for perhaps a better presentation or to show/hide the protein labels , which may prove useful for the visualization of relatively large networks . The reference section provides the references from which the interactions were extracted . The full source of resources used for the extraction of the interaction data can be accessed by going to the “References” section from top menu or going directly to the trypsNetDB . org/references . aspx webpage . The “Genome-wide data” section on the menu enables users to visualize the genome-wide data available for the queried proteins and their interacting partners suggested by the database . The supported genome-wide data in the current release of the database are categorized in three main groups of fractionation patterns , gene expression patterns , and phenotypic effects , with each containing sub-categories . Users can select one of the main categories ( indicated with a blue background ) to represent all related sub-categories at once or directly select the sub-categories . This part of the database is particularly useful for the validation of results obtained from interactome-related experiments ( such as affinity purifications ) by helping users to distinguish between direct binding partners of a protein from potentially spurious elements . For example , the fractionation heatmaps can be exploited to assess whether the potentially interacting proteins show similar fractionation patterns ( Fig 3 ) . Currently , the database provides fractionation patterns for whole-cell , mitochondrial-enriched , and cytosolic-enriched cell extracts that can be informative for the localization of previously unannotated proteins ( i . e . , mitochondrial proteins are expected to be identified in the mitochondrial-enriched fractions while depleted in the cytosolic-enriched sample ) . Fractionation patterns are also informative regarding the nature of the interactions . As described elsewhere [8] , glycerol gradient-based fractionation patterns can capture more transient interactions , while ion exchange-based fractionation favors more stable interactions due to the presence of a salt gradient . Finally , physically interacting proteins are expected to be involved in similar biological processes and , hence , show similar expression patterns and degrees of essentiality in each cell state , which can be easily assessed using the corresponding heatmaps . By going to the save option on the top of the network section , users can save the whole represented network or only the sub-networks that are supported by a specific experimental technique . It is also possible to save the enrichment analysis results and the annotations of genes , such as the description , transcript or protein characteristics ( length , weight , isoelectric point , and identified SNPs ) , and gene ontology . Users can also use the save query list option for later regeneration of the same results . The web application is developed based on the . Net framework 4 . 5 technology . To improve the performance , the statistical analysis modules ( including hypergeometric test , Wilcoxon-Mann-Whitney test , and Benjamini-Hochberg p-value adjustment procedure ) were implemented in C# and added as a library to the web application and the performance of the modules has been validated by comparing the results with those of MATLAB 2015b on multiple test sets to ensure accuracy . The network visualization is based on the cytoscape web library , which requires flash player for the representation of the network . All analyses are performed in real-time and a session for each user is ended after 1hr of inactivity . For high-performance , the database is implemented in Microsoft SQL Server 2012 .
Protein interaction maps remains one of the major resources for the functional annotation of proteins . Embedding other lines of information with these maps can help researchers gain insights regarding the molecular contexts of the proteins . Here , we introduce TrypsNetDB , a web tool to consolidate the current knowledge on the interactome of the trypanosomatid parasites and dynamically integrate them with a wealth of available orthogonal information . We are continuously working on expanding the core , literature-derived , protein interaction depository of the database . Future plans also include providing supports for the remaining trypanosomatid parasites and the inclusion of other genome-wide data . TrypsNetDB is an open source effort , and hence , the code and databases are available through the portal . Moreover , the interaction and fractionation data can be directly downloaded from the web interface using the provided links . | Methods to predict protein function based on sequences enable the rapid annotation of newly sequenced genomes . However , as most of these methods rely on homology-based approaches , non-conserved proteins in trypanosomatids remain elusive for annotation , rendering approximately half of the sequenced proteins uncharacterized . In this study , we developed a user friendly integrated database , TrypsNetDB , which fills multiple gaps in the field by depositing the current interactome knowledge on trypanosomatid proteins and combining this information with other available resources accompanied by related statistical analyses . The database allows automatic inter-species mapping of available data to better characterize the queried proteins in the species of interest . The database is built on fast and reliable ASP . Net framework and provides ( i ) a significant increase in the genome-wide functional annotation of trypanosomatid proteins , ( ii ) potential novel targets for therapeutics against trypanosomatids , and ( iii ) a robust methodology that can be adapted for the functional annotation of other non-model organisms . | [
"Abstract",
"Introduction",
"Program",
"description",
"and",
"methods",
"Conclusions",
"and",
"future",
"directions"
] | [
"protein",
"interactions",
"protein",
"interaction",
"networks",
"parasitic",
"protozoans",
"organisms",
"genomic",
"databases",
"trypanosoma",
"brucei",
"gambiense",
"protozoans",
"network",
"analysis",
"genome",
"analysis",
"sequence",
"motif",
"analysis",
"research",
"... | 2017 | TrypsNetDB: An integrated framework for the functional characterization of trypanosomatid proteins |
Cis-regulatory sequences are not always conserved across species . Divergence within cis-regulatory sequences may result from the evolution of species-specific patterns of gene expression or the flexible nature of the cis-regulatory code . The identification of functional divergence in cis-regulatory sequences is therefore important for both understanding the role of gene regulation in evolution and annotating regulatory elements . We have developed an evolutionary model to detect the loss of constraint on individual transcription factor binding sites ( TFBSs ) . We find that a significant fraction of functionally constrained binding sites have been lost in a lineage-specific manner among three closely related yeast species . Binding site loss has previously been explained by turnover , where the concurrent gain and loss of a binding site maintains gene regulation . We estimate that nearly half of all loss events cannot be explained by binding site turnover . Recreating the mutations that led to binding site loss confirms that these sequence changes affect gene expression in some cases . We also estimate that there is a high rate of binding site gain , as more than half of experimentally identified S . cerevisiae binding sites are not conserved across species . The frequent gain and loss of TFBSs implies that cis-regulatory sequences are labile and , in the absence of turnover , may contribute to species-specific patterns of gene expression .
Changes in gene regulation have been found in a wide range of species and can have a meaningful impact on cell and organismal phenotypes [1 , 2] . A significant fraction of regulatory variation can be attributed to changes in cis-regulatory sequences [3–7] . Changes in cis-regulatory sequences have been tracked to transcription factor binding sites ( TFBSs ) , insertion of transposable elements , and variation in tandem repeats , e . g . , [8–12] . Although changes in trans-acting factors are also important , e . g . , [13–15] , the molecular basis of changes in gene regulation will often require a dissection of cis-regulatory sequence evolution . A major challenge in studying the evolution of cis-regulatory sequences is translating divergence in cis-regulatory sequences to divergence in regulatory function . Although conservation of sequence is a strong indicator of conservation of function , cis-regulatory sequences that have maintained their regulatory function can diverge to the extent that they are unalignable [16–19] . On a finer scale , experimentally identified TFBSs are not always conserved across species [20–22] , even in cases when expression is known to be conserved [23] . The complex relationship between divergence in sequence and divergence in function [24] implies that the evolution of cis-regulatory sequences cannot be understood without investigating the evolution of individual TFBSs . The turnover of TFBSs provides a simple explanation for divergence in cis-regulatory sequences without a change in regulatory function . Under the binding site turnover model , the chance gain of a new binding site creates redundancy and can lead to loss of either the new or original site [21 , 23 , 25] . Evolutionary models suggest that many novel binding sites can be created by a stochastic mutational process and can potentially lead to the loss of existing sites [22 , 26–29] . Empirical evidence suggests that binding site turnover may be common . For example , the change in the position and orientation of binding sites within the even-skipped ( eve ) stripe 2 enhancer produces no change in embryonic patterns of expression between species , but chimeric enhancers composed from different species result in mis-regulation [23] . Furthermore , many experimentally identified binding sites have credible counterparts at close but not orthologous positions in other species [20–22] . Thus , the gain and loss of TFBSs is directly relevant to understanding conservation and divergence in cis-regulatory sequences in relation to their function . Models of TFBSs must account for sequence variations that have no affect on function or fitness [30 , 31] . Sequence variability within binding sites can arise as a consequence of a lack of specificity at certain positions or as a consequence of multiple sequences having the same binding energy . The specificity or binding probability of transcription factors for different DNA sequences has been modeled using both statistical mechanics [30] and information theory [32] . However , the relationship between binding probability and function or fitness is often not known . The simplest assumption is that both function and fitness are linearly related to the probability of being bound , which is approximately a step function of binding energy [30 , 33 , 34] . The distinction between sequences that can function as a binding site and sequences that cannot is critical to identifying the gain , loss , or turnover of TFBSs . The use of a cutoff , even one based on binding probability , is problematic when trying to classify sequences close to the cutoff [35] . One solution is to compare the likelihood of evolution under a model of neutral evolution to a model of a conserved binding site . Given a collection of known binding sites , the position-specific equilibrium base frequencies can be used to measure the strength of selection [36] and calculate the likelihood of evolution under a binding site model [28 , 37] . By combining models of neutral evolution with those for conserved binding sites , the frequency of conserved binding sites relative to those that have been gained or lost can be estimated [35] . Because the gain or loss of binding sites in nonfunctional sequences is common [22 , 26–29] , it is difficult to identify which gain or loss events are functional and affect fitness without additional data . One approach is to examine the gain and loss of experimentally identified binding sites . A previous study in Drosophila melanogaster found that 5% of Zeste binding sites , identified by chromatin immunoprecipitation , have been lost or gained across Drosophila species , based on deviations from a conserved binding site model [38] . However , nonfunctional sequences may often be bound without affecting gene expression [39] , and changes in gene expression may not always affect fitness [40 , 41] . A phylogenetic approach provides a means of identifying loss of functional binding sites based on significant conservation in some species but loss of constraint in others . This approach was used to identify cis-regulatory sequences around single-minded 2 ( SIM2 ) that were conserved in some but not all mammalian species [42] . Here , we have used a phylogenetic approach to examine the frequency at which functional TFBSs have been lost across the genomes of four Saccharomyces species . These species are sufficiently different that even the three closest species provide enough signal to identify individual binding sites by sequence conservation alone [43] . Using a probabilistic model of TFBS evolution [35] for 91 different transcription factors [44] , we found a substantial fraction of binding sites are not conserved between species , and that these sequence changes , at least in some cases , affect gene expression .
To identify semiconserved TFBSs , we used a probabilistic model of sequences evolving under a neutral and conserved binding site model . We define semiconserved sites as those that have been constrained along some lineages and unconstrained along others ( Figure 1 ) . Within this framework , semiconserved sites can be identified by their patterns of substitution rather than by their similarity to a binding site or a position weight matrix ( PWM ) representation of binding sites [45] . Additionally , semiconserved sites can be distinguished from conserved sites and neutrally evolving sequences by comparing the likelihood of a neutral model , a conserved binding site model , and a semiconserved model . The likelihood of a set of aligned sequences under a neutral model or a conserved binding site model is a function of the substitution rate under each model . Under a binding site model , the substitution rate depends on position-specific functional constraints imposed by the sequence specificity of a transcription factor . At equilibrium , the expected frequency of a nucleotide base is a function of the strength of selection on the base relative to the other bases [36] . Thus , the equilibrium frequency of bases from a collection of binding sites can be used to estimate the intensity of selection and the expected rate of substitution at each position [46] . To compare the likelihood of evolution under a neutral and conserved binding site model , we used synonymous sites to estimate the neutral substitution rate and PWMs to estimate the equilibrium base frequencies within binding sites and derive position-specific substitution rates ( see Methods ) . The likelihood under a semiconserved model depends on which lineages have evolved under a neutral model and which have evolved under a binding site model . The semiconserved model can , in theory , detect both the loss and gain of binding sites . However , constraint on only a single lineage is typically indistinguishable from neutral evolution . Thus , we limited our analysis to loss of constraint on a single lineage and we did not consider loss events on the outgroup lineage . Since the lineage and time at which loss of constraint occurred is unknown , we calculated the likelihood under the semiconserved model by integrating over a large number of loss events evenly distributed over all lineages excluding the outgroup lineage , an approximation of the method used by Mustonen and Lassig [35] . To estimate the frequency of semiconserved relative to conserved binding sites , we used 91 TFBS models [44] and 1 . 7 megabases of noncoding sequences from 3 , 761 multiple sequence alignments of S . cerevisiae , S . paradoxus , S . mikatae , and S . bayanus [47] . Rather than test every position in the genome alignments , we calculated the likelihood under each model for the 2 , 000 positions with the highest-scoring sequence match to each binding site model in any two of the four species ( see Methods ) . We used expectation maximization to obtain an overall estimate of the frequency of sites evolving under each model . We found that 55% of the sites are best explained by the conserved binding site model , 31% are best explained by the semiconserved model , and 14% by the neutral model . The frequency of neutral sites is arbitrary since we did not test all positions within the alignments . Of the non-neutral sites , one-third are better explained by a model that allows for loss of constraint along one lineage . However , this estimate includes many sites that are reasonably explained by all three models . Sequences that don't provide a close fit to the conserved or semiconserved model may be evolving under a similar , yet unknown , model and may be incorrectly annotated as a semiconserved binding site . Figure 2A shows the posterior probabilities for 2 , 000 putative Rox1 sites present in the yeast genome alignments . Because the posterior probabilities sum to one , sites with a high likelihood under the semiconserved model but not the neutral or conserved model are shown in the bottom left corner , and sites with a high likelihood under the conserved model but not the neutral of the semiconserved model are shown in the upper left corner . The distribution of probabilities suggests that quite a few sites are equally well explained by each model . To estimate our confidence in identifying individual sites that have evolved under a conserved or semiconserved model , and to eliminate sequences that may be evolving under a similar model , we generated null distributions for each model using computer simulations . Figure 2B shows the posterior probabilities for 2 , 000 sites simulated under a neutral model and 2 , 000 sites simulated under a model of a conserved Rox1 binding site . Three cutoffs were used to generate the high-confidence set of conserved and semiconserved sites ( Figure 2B ) . The first cutoff delineates sites with a low probability under the neutral model ( p ( neutral ) < 0 . 005 ) . The second and third cutoffs delineate sites with a high probability under the conserved model and the semiconserved model , respectively . The second cutoff is set such that fewer than 1% of neutral sites show a higher likelihood under the conserved model . The third cutoff is set such that fewer than 1% of conserved sites show a higher likelihood under the semiconserved model . Out of 2 , 000 putative Rox1 sites , 292 were inconsistent with a neutral model ( cutoff 1 , Figure 2A ) . Of these 292 sites , 242 sites were defined as conserved ( cutoff 2 ) and 11 as semiconserved ( cutoff 3 ) . Out of 2 , 000 neutral simulations , two were defined as conserved and three were defined as semiconserved based on our cutoffs . Out of 2 , 000 conserved binding site simulations , 1% passed the semiconserved cutoff , suggesting that 292*1% ≈ 3 of the semiconserved sites are false positives . This data translates into a false discovery rate of 2/242 ( 1% ) for conserved sites and 6/11 ( 54% ) for semiconserved sites . The false discovery rates indicate that our cutoffs do not exactly produce a high-confidence set of semiconserved sites . However , simulations of semiconserved sites show the power to detect semiconserved Rox1 sites is only 16 . 4% , and increasing the stringency would reduce the power further ( Figure 2C ) . Using 91 TFBS models [44 , 48 , 49] , we estimated the fraction of semiconserved sites for each . In total , we found 19 , 264 sites showed evidence of non-neutral evolution ( p < 0 . 005 under the neutral model ) . Of these non-neutral sites , we classified 15 , 399 as conserved ( p > 0 . 99 for the conserved model ) , and 982 as semiconserved model ( p > 0 . 99 for the semiconserved model ) ( Table 1 ) . In total , of the significantly conserved or semiconserved binding sites , 6 . 0% have been lost in a lineage specific manner . Semiconserved binding sites were identified for 85 out of 91 binding site models , and more than five loss events were found for 60 of the 91 models . To estimate the rate of false positive classification of conserved and semiconserved sites , we simulated 2 , 000 neutral and 2 , 000 conserved binding sites for each model . Classifying these simulated sequences , we found 224 neutral sequences passed our cutoffs for a conserved site and 242 neutral sites passed our semiconserved cutoffs . Thus , the rate of falsely classified conserved sites is just over 1% ( 224/15 , 399 ) . By definition , 1% of the simulated conserved sites passed the semiconserved cutoff . Thus , the overall false discovery rate for semiconserved sites is 44% ( 19 , 264 * 0 . 01 + 224 ) /982 . The classification of sequences into conserved and semiconserved sites supposes that all sequences bound by the same protein evolve under the same functional constraints . However , for any given transcription factor , there may be certain sites in the genome that are selected for high binding energy and other sites that are selected for lower binding energy [33 , 35 , 50] . Selection for low-energy sites may produce the appearance of semiconserved sites if analyzed using a model based on high-energy sites . To investigate whether semiconserved sites may be low-energy binding sites , we examined the binding energies of conserved and semiconserved sites . We used the likelihood ratio score of a sequence under a binding site model compared with a model of background sequences as a proxy for binding energy [31] . The distribution of scores shows that semiconserved binding sites tend to have higher binding energy than the completely conserved sites on the lineages in which they have been conserved . In the lineage showing loss of constraint , the binding energies are much closer to background sequences ( Figure 3 ) . Additionally , the substitution rate within semiconserved sites is indistinguishable from that of conserved sites , excluding those lineages showing loss of constraint ( Table 2 ) . These comparisons suggest that semiconserved sites cannot be explained by a class of low-energy sites . Two models can explain the lineage-specific loss of TFBSs . First , some species may experience new environments where certain regulatory elements are not needed , or are selected against , resulting in a change in gene regulation . Second , the gain of one or more redundant binding sites within a promoter enables the loss of a previously constrained site ( Figure 1D ) . Under the second model , the turnover of function from one binding site to another conserves the regulatory control but enables divergence within regulatory sequences . The binding site turnover model predicts that binding site loss will be accompanied by the gain of a site elsewhere in the promoter . We tested this prediction by looking for the presence of a species-specific binding site for the same transcription factor in the promoter showing loss of constraint . We defined species-specific binding sites as a sequence that matches a PWM in one species , but whose orthologous sequences do not match the same weight matrix . To define a match to a PWM , we used a log-odds score cutoff from the tenth percentile score of conserved binding sites for each binding site model . Using this cutoff , 57% ( 513/894 ) of the species-specific loss events can be explained by turnover ( Table 1 ) . In comparison , species-specific sites are present within 50% of promoters with conserved sites and 47% of promoters with semiconserved sites , excluding lineages with loss . Using a more stringent cutoff score derived from information theory [45] , 38% of the loss events can be explained by turnover . Binding site turnover is not due to any one lineage or binding site model . The rate of turnover is similar across lineages , with 50% of sites showing turnover in S . cerevisiae , 55% in S . paradoxus , and 60% in S . mikatae . Although the rate of turnover varies across binding site models , most of this variation can be explained by the information content of the models and the size of the promoter sequences within which semiconserved sequences lie , consistent with previous work [28] . Natural selection may also result in lineage-specific loss of TFBSs . If the fitness effects of binding sites differ between species , bind sites may be lost without consequence or they may be selected against . However , it is also possible that semiconserved sites arise from compensatory changes that are more complicated than those described by a simple binding site turnover model . For example , binding site turnover may also occur between sites bound by different but functionally related transcription factors . Distinguishing between these two possibilities is not easy . If some but not all species have undergone a substantial shift in selective pressures , binding site loss may show high rates on specific lineages . In contrast , if binding site loss is the result of turnover , loss should be a simple function of sequence divergence . The number of loss events on each lineage is heterogeneous ( Table 1 ) . Scaled by the synonymous substitution rate along each lineage , S . mikatae shows the greatest amount of loss , 66% of the loss events but only 40% of the total evolutionary distance , and S . paradoxus shows the least , 3% of the loss events but 16% of the evolutionary distance . However , simulations of semiconserved sites with loss events evenly distributed over the tree shows that the power of detecting binding site loss is the lowest on the shortest lineages , since these lineages have the fewest informative substitutions . One way to control for the confounding effects of power is to identify binding sites that show lineage-specific rates of loss that differ from the average lineage-specific rate across all binding site models . Using the average rates of lineage-specific loss across all binding sites as a control ( Table 1 ) , we tested 29 binding site models with at least ten loss events for a heterogeneous distribution of binding site loss across lineages . We found significant heterogeneity in the loss of both Spt23 and Rlr1 binding sites ( X2 , 3 d . f . , p = 3 × 10−7 for Spt23 and p = 4 × 10−11 for Rlr1 ) . Spt23p stimulates Ty1 transposition and is a suppressor of Ty1-induced promoter mutations [51] . For Spt23 sites , the largest amount of loss was found on the lineage leading to S . paradoxus ( 14 loss events observed , 3 . 5 expected ) . Rlr1 is involved in transcription associated hyper-recombination between direct repeats [52] . For Rlr1 , the largest amount of loss was found on the lineage leading to the ancestor of S . cerevisiae and S . paradoxus ( 38 loss events observed , 14 . 6 expected ) . The lineage-specific rate of loss of Spt23 and Rlr1 sites suggests that the loss of these sites may not have been a stochastic process . In the absence of binding site turnover , the semiconserved model predicts that the substitutions resulting in binding site loss should cause changes in gene expression . To experimentally determine whether semiconserved sites are functional and whether substitutions predicted to cause binding site loss are functional , we recreated the loss of three Rox1 , two Ndt80 , and six Msn2/4 semiconserved binding sites . These semiconserved sites were picked from 11 , 14 , and 27 semiconserved binding sites predicted using the Rox1 , Ndt80 , and Msn2/4 binding sites models , respectively . For each semiconserved site , we used a beta-galactosidase reporter construct to compare the expression of the wild-type S . cerevisiae promoter with a mutated S . cerevisiae promoter containing the same substitutions predicted to cause change of function ( Figure 4 ) . Expression was measured in two strains of S . cerevisiae , one with and one without the transcription factor predicted to bind the site of interest . Mutations in five of the 11 semiconserved binding sites affected levels of gene expression ( Table 3 ) . If these changes in expression are caused by the transcription factor predicted to bind the site , they should be absent in strains lacking the transcription factor . Using transcription factor deletion strains , we found that in only three of the five cases were these effects dependent on the presence of the transcription factor predicted to bind the site . Out of three semiconserved Rox1 binding sites , the site in the SUT1 promoter showed a Rox1-dependent effect on gene expression . The three substitutions resulted in a 1 . 6-fold increase in gene expression , consistent with Rox1 being a transcriptional repressor in the presence of oxygen [53] . Both of the semiconserved Ndt80 binding sites produced a significant effect on gene expression ( Table 3 ) . However , only in the HST4 promoter is the effect dependent on Ndt80 . The two substitutions in the HST4 promoter led to a 1 . 7-fold decrease in gene expression during sporulation , consistent with Ndt80′s role as activating the middle sporulation genes [54] . In the NAM8 promoter , a single substitution caused a 3-fold increase in expression during vegetative growth , independent of Ndt80 . Out of the six semiconserved Msn2/4 binding sites , the substitutions affected expression in two cases . Yet , of the two functional sites , only the one in the MDH1 promoter affected expression in an Msn2–Msn4 double mutant ( Table 3 ) . Interestingly , this effect was only present during nitrogen starvation and not during heat shock . At equilibrium , the rate of binding site gain should be comparable to that of binding site loss . We previously showed that two Ndt80 binding sites , which show no conservation in other species , affected gene expression in S . cerevisiae [43] . Although the gain of a binding site that affects gene expression levels may be inconsequential to fitness , and thus susceptible to loss , the frequency at which functional binding sites are gained is relevant to understanding the evolution of gene regulation . To estimate the rate of binding site gain across multiple transcription factors , we obtained a list of documented binding sites from the Yeastract database [55] . Because this database does not contain exact coordinates for each binding site , but rather transcription factor–promoter pairings , we limited our analysis to the 654 binding sites for 61 transcription factors where there was only a single high-scoring sequence match to the PWM in the promoter of interest . For each binding site , we tested its conservation across the four Saccharomyces species . We found that 303 ( 46 . 3% ) of the Yeastract sites fit the conserved model , and seven ( 1 . 1% ) fit the semiconserved model . Thus , a substantial fraction of experimentally identified binding sites appear to be species-specific or only weakly conserved across species , implying that binding site gain may be common .
The frequency of experimentally identified binding sites that are not conserved across species suggests a high rate of binding site gain . We found that more than half of the binding sites extracted from the Yeastract database [55] are not conserved . This is consistent with studies in other organisms . Between 30% and 50% of experimentally identified binding sites lie outside of conserved blocks in Drosophila [20] , 40% of human and mouse TFBSs are species-specific [22] , 5% of Zeste binding sites are not conserved among closely related Drosophila species [38] , and 5% of CRP binding sites show presence and absence at orthologous positions in two bacterial genomes [35] . However , the biological relevance of these unconserved sites is not always known . Sites that are bound and affect gene expression may in some cases be lost without any fitness or downstream phenotypic consequences , except for a change in gene expression . In comparison , a binding site that has been conserved in some species but lost in others suggests that the site is relevant to fitness , at least in those species in which it was conserved . The frequency of binding site loss may be quite high , but is difficult to estimate . Using expectation maximization , we estimated that one-third of all non-neutral sites are no longer constrained on some lineages . However , this estimate does not account for sequences that may have evolved under functional constraints other than the binding site model being tested . Using a number of statistical cutoffs to eliminate ambiguous sites , we found that 6% of the high-confidence binding sites fit the semiconserved model . This is similar to other estimates of the frequency of functional binding sites that are not entirely conserved across species [35 , 38] . Although some of these sites may be false positives , the true number of semiconserved sites could be higher , given that we estimated our power to detect semiconserved sites to be low , less than 20% for most models . In the absence of binding site turnover , binding site loss results in species-specific changes in gene regulation . This model predicts that changes in gene expression should result from either making substitutions that result in loss in the species with a conserved site , or from making substitutions that recreate the binding site in the species showing loss . We tested the former of these two predictions using 11 different predictions of binding site loss . In three cases , we found that the substitutions predicted to result in loss of function altered the expression of the downstream gene . Although suggestive , these experiments do not address whether the substitutions that occurred on the lineage showing loss resulted in a species-specific change in gene regulation . The eight of 11 semiconserved sites that showed no affect on gene expression are difficult to interpret . One explanation is that the semiconserved sites only affect gene expression under specific environmental conditions . Although possible , the gene expression assays were carried out under conditions where the semiconserved sites were likely to function . Another explanation is that our assays were not sensitive enough to detect small changes in gene expression . Finally , the predictions rest on the false positive rate of the model as well as on its assumptions . While it is difficult to distinguish between these possibilities , several pieces of evidence suggest that the assumptions of the model may not always be correct . Our predictions of binding site loss rest on a number of assumptions . The main assumptions are that the alignments are correct , the binding site models are correct , and that sequences that appear to be semiconserved binding sites are not functionally constrained for some other reason . We discuss each of these assumptions separately . Although functional divergence in cis-regulatory sequences may be common , in relatively few cases have the nucleotide substitutions been identified [59] . TFBSs provide a useful starting point to dissecting sequence divergence that underlies regulatory divergence . The semiconserved model we have used in this analysis provides an efficient way to identify loss of constraint on a putative binding site sequence . Although several good candidate loss events were identified , there is a considerable false positive and false negative rate associated with the approach . Additional comparative information should help eliminate false positives , and methods that account for uncertainty in the binding site model should improve our ability to reliably detect functional divergence in cis-regulatory sequences .
For each model , we assume that nucleotide sequences are evolving under a discrete-state , continuous-time Markov process , positions within an alignment evolve independently of one another , and the substitution rate is a product of the population size , N , mutation rate , μ , fixation probability , f , and time , t , measured in generations . We also assume that the mutational process is the same under each model and is governed by five parameters [60]: four parameters for the equilibrium nucleotide frequencies ( πa , πc , πg , πt ) and one parameter for the rate of transitions relative to transversions ( κ ) . The probability of fixation is different between the models . Under the neutral model , the probability of fixation is the same for all mutations . Under the binding site model , the relative probability of fixation between any two bases is: where s is the selective advantage of base y relative to base x [61] . The strength of selection can be estimated from the equilibrium base frequencies [36 , 62] . Given a collection of sites evolving under the same model , at equilibrium , the flux from base x to base y is equal to the flux from base y to x: where πx is the equilibrium frequency of base x , μ is the mutation rate , and f is the fixation probability . Using the approximation of Equation 1 , which assumes Ns > 1 , and Equation 2 , the equilibrium base frequencies are a simple function of the relative strength of selection and mutation: Substituting Equation 3 into Equation 1 , the probability of fixation is: In the binding site model , the fixation probability is position-specific and derived from PWMs , as described below . Assuming the effective population size is constant , no estimate of N is needed since it is the same across all positions and all types of nucleotide changes . We calculated the likelihood of the data under the neutral and conserved binding site model using transition probabilities derived from the expected rate of substitution under each model and using the pruning algorithm to integrate over all possible ancestral states [63] . To estimate the expected rate of substitution , we estimated κ from substitutions in synonymous sites in coding sequences ( κ = 4 ) , the π parameters from the genome-wide nucleotide frequencies ( A = 0 . 3 , G = 0 . 2 , C = 0 . 2 , T = 0 . 3 ) for the neutral model and from PWMs for each TFBS model . We estimated the mutation rate and time , together , for each branch of the phylogeny from synonymous sites using PAML [64] . Given these branch-specific substitution rates , we calculated the transition probability under each model by exponentiating the rate matrix , P = eQ , where Q is a matrix of substitution rates of the form 2Nμπft . For a pair of sequences , x and y , the likelihood of an aligned binding site , S , of width W , is given by: Here , a represents the nucleotide in the ancestral sequence A . TAX is the branch length from the ancestor to species X . QiaX is the substitution rate from base a to base X in position i . Q can be either the neutral model of evolution ( in which case it is position-independent ) , or the binding site model . υa is the frequency of base a in ancestral sequence . For neutral sequence , this is the genome average frequency , πa . For the binding site model , this is the frequency of a in position i of the PWM . Equation 5 can be expanded to multiple sequences by recursively calculating the left and right branches of each node in the phylogenetic tree starting at the root [63] . To calculate the likelihood under the semiconserved model , we integrated over many loss events evenly distributed across the entire tree , excluding the outgroup . By re-rooting the tree at the time-point , t , where constraint was lost , we split the tree into two subtrees , with one subtree containing all sequences preceding t , and the other subtree with all sequences following t . The likelihood of the left and right subtrees was then calculated under the binding site model and the neutral model using the pruning algorithm and Equation 5 . Thus , the likelihood under the semiconserved model is: where D is the total evolutionary distance , S is the aligned binding site , Tt is the portion of the tree evolving under the binding site specific model of evolution , T − Tt is the neutrally evolving portion of the tree . Because very recent loss events are indistinguishable from the conserved binding site model , we do not test for loss events occurring within 0 . 1 substitutions per site of the extant species . We used the maximum-likelihood estimate of the location of the loss event to determine on which branch the loss of constraint occurred . Pseudocode can be found in Protocol S1 . To estimate the fraction of sites that are evolving under a semiconserved model of evolution , we used a maximum-likelihood approach . Using expectation maximization , we maximized the likelihood equation: p ( conserved | sitei ) and p ( neutral | sitei ) were calculated using the pruning algorithm and Equation 5 . p ( semiconserved | sitei ) was calculated using Equation 6 . p ( conserved ) , p ( semiconserved ) , and p ( neutral ) are the free parameters that were maximized . To distinguish between the three models , we compared the posterior probability of each model . While the maximum-likelihood estimates suggested that the probabilities of the three models are unequal , we used flat priors for simplicity . The choice of priors did change the overall annotations slightly , but the general conclusions are unchanged . Computer simulations of neutral and conserved sequences were used to set statistical cutoffs for distinguishing each model and to estimate the power of detecting binding site loss . For each simulation , we evolved a sequence from the root of the tree to each node/tip using the transition probabilities specific to each model . For both simulations , we generated sequence at the root from the nucleotide frequencies defined by the PWM . We used 10 , 000 neutral simulations to generate the neutral cutoff ( #1 in Figure 2B ) , such that less than 0 . 5% of sites show a lower posteriori probability under the neutral model . The same data were used to generate the conserved cutoff ( #2 in Figure 2B ) , such that less than 1% of neutral sites show a higher posteriori probability under the conserved model . We used 5 , 000 conserved binding site simulations for each transcription factor to generate the semiconserved cutoff ( #3 in Figure 2B ) , such that less than 1% of sites show a lower probability under the conserved model . To control the false discovery rate and computational time , we tested only the 2 , 000 highest-scoring binding sites for each transcription factor . To identify these sites , we ranked each putative binding site by the sum of the two highest-scoring sequences from the four species examined by their log-odds score , see below . The choice of 2 , 000 sites is arbitrary , but as most transcription factors are expected to regulate fewer than a few hundred genes , this should not exclude any functional binding sites from our analysis . A summary table of the data for all 91 transcription factors can be found in Table S1 . The genomic coordinates of all conserved and semiconserved binding sites are provided in Table S2 . Alignments . ClustalW intergenic sequence alignments of S . cerevisiae , S . paradoxus , S . mikatae , and S . bayanus [47] were filtered to remove any alignments containing greater than 50% insertions or deletions in any one sequence or those containing greater than 20% missing data ( N and . characters ) . After applying these filters , global alignments of 3 , 761 intergenic sequences from four species were used in all subsequent analysis . 1 , 539 coding sequence alignments were used to estimate the synonymous and nonsynonymous substitution rates [64] . To account for insertion or deletion events within aligned binding sites , we generated local realignments by using the highest-scoring binding site in each species from the binding site and ±5 bp of it , excluding gaps . TFBS models . We used the TFBS models defined by Harbison et al . [44] , with the addition of a model for Ndt80 [48] and CSRE [49] , as these well-studied motifs were not included . We filtered the dataset to remove dubious or redundant motifs , and used 91 out of 104 reported binding site models ( see Table S1 ) . Defining TFBS turnover . We define binding site turnover as the presence of a species-specific binding site in the promoter of the species showing loss . To identify species-specific binding sites , we used the log-likelihood ratio score of the sequence given a PWM: where πib is the frequency of base b at position i of the binding site as defined by the PWM , and ρb is the genomic frequency of base b , and W is the width of the binding site . To determine the cutoff score for a sequence match to the PWM , we used the distribution of scores in sites identified as significantly conserved . For each transcription factor , we enumerated the scores from all four species for each conserved binding site and used the tenth percentile of these scores to define a match to the PWM . To estimate the expected number of turnover events , we calculated the percentage of promoters containing a species-specific binding site for each transcription factor . We also used the default cutoff score of the Patser program [45] for comparison . Beta-galactosidase activity driven by both a wild-type and mutant promoter sequence was measured to determine the effect of the binding-site loss on gene expression . For each putative loss event , the entire S . cerevisiae intergenic sequence was cloned by PCR with gap-repair or restriction digests into the YEp357r yeast–bacteria shuttle vector [65] . Mutations were made in the binding site to mimic the substitutions that occurred between species using stitching-PCR and were confirmed by sequencing . The constructs were transformed into the S . cerevisiae strain BY4743 or the appropriate homozygous deletion strain , obtained from the yeast deletion collection , for the transcription factor of interest [41] . The msn2Δmsn4Δ double-deletion strain was generated from a cross between the two single-deletion strains and confirmed by PCR . To measure gene expression driven by either the S . cerevisiae binding site or the mutated binding site , yeast cultures were grown overnight in complete minimal medium minus uracil and diluted to a starting OD600 of 0 . 05 . Each construct was measured in selective media during mid-log phase growth . The Ndt80 binding sites were also measured after 10 h in 1% potassium acetate . The Msn2/4 binding sites were also measured following a heat shock of 1 h at 37 °C , or 8 h in media with no nitrogen source . The 11 sites tested were selected before the final statistical tests were applied . As a consequence , two of the 6 Msn2/4 binding sites , in YAL008W and GSP2 , had a posteriori probability under the neutral model of 0 . 005 < p < 0 . 01 . We downloaded the set of documented S . cerevisiae binding sites from the Yeastract database [55] . Because only transcription factor promoter pairings are reported ( e . g . , transcription factor X regulates gene Y ) , we limited the analysis to the 654 promoters with only a single high-quality match ( greater than the 25th percentile of the log-odds scores of the conserved binding sites ) to the transcription factor's binding site . | Research in the field of molecular evolution is focused on understanding the genetic basis of functional differences between species . Protein coding sequences have traditionally been the focus of these studies , as the genetic code enables a detailed study of the strength of selection acting on amino acid sequences . However , from the earliest cross-species sequence comparisons , it was clear that protein sequences among closely related species are too similar to explain the observed phenotypic diversity . This led to the hypothesis that the evolution of gene regulation has played a key role in generating diversity between species . The availability of numerous complete genome sequences has made it possible to begin testing this hypothesis . In this work , the authors use an evolutionary model to identify functional divergence within transcription factor binding sites , the core functional elements involved in gene regulation . Applying this model to the baker's yeast , Saccharomyces cerevisiae , and its three closest relatives , the authors find that a substantial fraction of the ancestral binding sites have been lost in a species-specific manner . In some cases the loss of the binding site creates gene expression differences that may be indicative of species-specific changes in gene regulation . This work provides a useful computational framework that will allow further study of the conservation of cis-regulatory sequences and their role in molecular evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"yeast",
"and",
"fungi",
"eukaryotes",
"computational",
"biology",
"evolutionary",
"biology",
"genetics",
"and",
"genomics",
"saccharomyces"
] | 2007 | Frequent Gain and Loss of Functional Transcription Factor Binding Sites |
Influenza viruses have been responsible for large losses of lives around the world and continue to present a great public health challenge . Antigenic characterization based on hemagglutination inhibition ( HI ) assay is one of the routine procedures for influenza vaccine strain selection . However , HI assay is only a crude experiment reflecting the antigenic correlations among testing antigens ( viruses ) and reference antisera ( antibodies ) . Moreover , antigenic characterization is usually based on more than one HI dataset . The combination of multiple datasets results in an incomplete HI matrix with many unobserved entries . This paper proposes a new computational framework for constructing an influenza antigenic cartography from this incomplete matrix , which we refer to as Matrix Completion-Multidimensional Scaling ( MC-MDS ) . In this approach , we first reconstruct the HI matrices with viruses and antibodies using low-rank matrix completion , and then generate the two-dimensional antigenic cartography using multidimensional scaling . Moreover , for influenza HI tables with herd immunity effect ( such as those from Human influenza viruses ) , we propose a temporal model to reduce the inherent temporal bias of HI tables caused by herd immunity . By applying our method in HI datasets containing H3N2 influenza A viruses isolated from 1968 to 2003 , we identified eleven clusters of antigenic variants , representing all major antigenic drift events in these 36 years . Our results showed that both the completed HI matrix and the antigenic cartography obtained via MC-MDS are useful in identifying influenza antigenic variants and thus can be used to facilitate influenza vaccine strain selection . The webserver is available at http://sysbio . cvm . msstate . edu/AntigenMap .
An influenza virus is a negative-stranded RNA virus that belongs to the Orthomyxoviridae family . There are three serotypes , A , B , and C , of which B and C are reported to infect mammals only . The influenza A viruses have genomic segments ( segment ) with varying lengths from about to nucleotides which encode at least proteins: PB2 by segment , PB1 and PB1-F2 by , PA by , haemagglutinin ( HA ) by , nucleoprotein ( NP ) by , neuraminidase ( NA ) by , matrix protein M1 and M2 by , and nonstructural protein NS1 and NS2 by . Among these proteins , the surface proteins HA and NA are involved in virus attachment and cell fusion . Both HA and NA are the primary targets for host immune systems . The serotypes of influenza A viruses are based on HA and NA subtypes . To date , HA and NA subtypes have been reported in influenza A viruses [1] . For instance , H1N1 influenza A virus is named since it has HA and NA recognized by HA subtype and NA subtype antibodies , respectively . Influenza B viruses have segments while Influenza C has segments . There is not yet an HA-NA nomenclature system in Influenza B and C viruses . The peak influenza season in the northern hemisphere is from January to April every year . More than hospitalizations and deaths are caused by influenza in the United States each year [2] , [3] . The influenza A virus may cause a pandemic disaster that will impact multiple continents . In the 20th century , three influenza A pandemics occurred in 1918 , 1957 , and 1968 , respectively [4] , [5] . More than million people were killed in the 1918 influenza pandemic , which was caused by the H1N1 influenza A virus . This influenza pandemic shortened global life expectancy by more than years . During March and early April , a new H1N1 influenza A virus epidemic was detected in Mexico and the United States [6] , and the virus spread rapidly through human-to-human transmission , resulting in WHO declaring a pandemic , which was the first influenza pandemic in the past years . This virus was estimated to cause about million infections and deaths solely in United States through Jan 14 , 2010 ( www . cdc . gov ) . If we consider all cases in five continents , the numbers will become significantly larger . In the United States , vaccination is the primary option for reducing the effects of influenza . The seasonal influenza vaccines used in the past decades include three viral components: H1N1 influenza A virus , H3N2 influenza A virus , and influenza B virus . In an effective vaccination program , vaccine strain selection will be the most important step since the highest protection could be achieved only if there is an identical antigenic match of the vaccine and epidemic virus HA and NA antigens , especially HA , which is the primary target of human immune system . However , as an RNA virus , influenza A virus has rapid mutations in these two proteins , and such mutations can cause a change of antigenicity , thus making vaccines ineffective . Mutations in HA and NA are also referred as antigenic drift . Immunological tests , such as hemagglutination inhibition ( HI ) assay , enzyme-linked immunosorbent assay ( ELISA ) , and microneutralization assay , have been utilized to identify antigenic variants among the circulating influenza strains . Among these assays , HI , has been one of the routine procedures in influenza vaccine strain selection . HI assay is an experiment to measure how a testing influenza antigen ( virus ) and a reference antiserum ( antibody ) react . The antibody is usually diluted in fold first and then diluted in powers of . Thus , the titre from HI assay will be , . The larger the is , the more closely the testing antigens match the reference antigens , for which the reference antisera are generated . Usually a number smaller than is considered as a low reaction between antigen and antibody . In many cases , HI experiments are used to measure the antigenic distance between two testing antigens through their immunological reactions to the same reference antiserum . For instance , if one testing antigen is a high reactor for the reference antiserum ( e . g . with a titre of ) while another testing antigen is a low reactor ( e . g . with a titre of ) . The antigenic distance could be approximately units , which is . In reality , the antigenic distances are usually measured by a set of reference antisera , thus the calculation is much more complicated . Such measurements from HI data are generally used to determine the antigenic distances between testing antigens . In a typical influenza HI assay , generally less than reference antisera are used but the number of test antigens can be more than . However , interpretations of HI results are not straightforward due to the following two challenges: ( 1 ) HI assay only shows the indirect relationship between antigens and antisera since each value reflects a reaction from antigen , red blood cell ( RBC ) , and antibody . Many variables from RBC and antibody will interfere the HI titres; ( 2 ) it is not be possible to perform HI for all pairs of antigen and antisera reactions . Thus , the resulting HI table is generally incomplete , and the percentage of missing data could be up to . By applying the metric multidimensional scaling method ( MDS ) to reduce the shape space into less than three dimensions , Lapedes and Farber [7] showed a linear correlation between logarithm values of HI titers and the space distances between influenza antigens . Based on this method , Smith et al . [8] constructed influenza cartography to visualize the distances among influenza antigens from HI tables by further developing the metric MDS method . Their method assumes that antigens and antibodies are mapped into the same low-dimensional space , and their interactions are the distances between the embedded points . However , in our implementation of their algorithm , the resulting influenza cartography depends on the initial values selected , and thus may not be stable . Moreover , this method results in cartographies in which global distances may contain relatively large errors . This is because the algorithm does not incorporate temporal modeling to reduce the inherent temporal bias in HI tables . The temporal bias is caused by the fact that HI table entries are not missing uniformly at random , and off diagonal entries are more likely to be missing or become low reactors ( Figure 1 ) . The underlying biological reason for this bias can be explained by the herd immunity effect , where influenza antigens evolve rapidly under the accumulating immune pressures of human population [9] . A more detailed illustration of this phenomenon will be given later . The goal of this paper is to present a computational framework for influenza cartography construction which we call Matrix Completion-Multidimensional Scaling ( MC-MDS ) . An important aspect of this framework is that temporal modeling can be easily incorporated , which as we shall show , is useful for dealing with HI tables with herd immunity induced temporal bias . Our framework includes two integrated steps: ( 1 ) a low rank matrix completion algorithm is first employed to fill in the entries of the HI matrix; ( 2 ) a MDS algorithm is utilized to map the antigens ( or similarly , antibodies ) into a two dimensional space for visualization . Our approach explicitly separates the visualization ( cartography ) step from the matrix completion step , making it easier to incorporate temporal models . Our experience shows that while temporal modeling is beneficial in both steps , it is less important in the first step , for which we may simply employ a sliding window approach; however it is more essential in the second step , for which we propose a more complex herd-immunity temporal regularization model as described in the Materials and Methods section . The reason for the difference is that the inherent temporal bias tends to give rise to incorrect global distances if not handled explicitly , and thus affect the 2D cartography process more significantly . The two step procedure in our approach is thus flexible in the first step , where we can simply use a standard low rank matrix completion algorithm . On the other hand , we have to pay special attention to temporal modeling in the second step , which is essential for accurate cartography construction . Both simulation and a practical application in H3N2 influenza A viruses demonstrate that this method is able to overcome some limitations in the original metric MDS method of [8] and it results in better influenza antigenic cartographies from HI data . Therefore the proposed framework can potentially facilitate more accurate interpretation of HI data in influenza surveillance as well as more accurate identification of influenza antigenic variants . Both are essential for influenza vaccine strain selection .
In this work we are specifically interested in HI datasets existing accumulating original , such as the immunological datasets of human origin . In a typical HI dataset , three types of data entries are present: Type I , a regular HI titre; Type II ( low reactors ) , the value is defined as “less than a threshold” , e . g . and this threshold is caused by the lower bound experimental limit in HI assays indicating a weak ( or low ) immunological reaction between a testing antigen ( virus ) and an antiserum ( antibody ) ; Type III , missing values . A major characteristic of HI dataset is that the distributions of type I , type II , and type III data are not random . Specifically , if we arrange both antigens and antibodies in a HI matrix according to time , then there is a banded structure , where most Type I data appear very close to the diagonal of the matrix; Type II data tend to be slightly off diagonal , while Type III data are more likely to occur in matrix entries that are significantly off diagonal ( Figure 1 ) . This data characteristic introduces a “temporal bias” concerning the data distribution ( in comparison to uniformly random distribution ) that needs to be corrected . As we will show , if the problem is not handled appropriately , then inaccurate result will be produced . This is because classical methods assume uniformly random data distribution , which does not take the temporal bias effect into consideration . Our paper shows that temporal modeling , which reduces the data distribution bias in HI tables , is important in HI based influenza cartography . The specific benchmark dataset used in our study includes entries , representing of all table entries ( Figure 2 ) . Among these entries , ( ) are Type II values ( that is , they are recorded as ) with . For algorithmic comparison purposes , we also include results on a simulation dataset with ground truth , which is generated according to characteristics of real HI datasets . As pointed out above , most Type I data are located across the diagonal line of the HI matrix , which significantly deviates from the “missing uniformly at random” assumption in classical matrix completion . In order to reduce this bias , we adopt a sliding window approach where each low rank matrix completion will be performed in a HI sub-matrix , which has fewer amount of Type II and Type III data that more closely satisfy the “missing uniformly at random” assumption . The remaining entries that are not covered by the ( sliding window ) sub-matrices can be filled with a global matrix completion algorithm – those entries will be predicted with less accuracy due to the banded-structure of the HI data that violates the “missing uniformly at random” assumption . The windows are based on the temporal spans of influenza A viruses . In order to complete the entire matrix , the algorithm will slide yearly along with both the dimensions of antigens and antisera to ensure the time difference between all antigens and antisera are within a certain window size . In order to obtain an optimal window size and best rank in matrix competition , we tested six different sizes , including , , , , , and , and ranks to . A -fold cross validation suggested that the time frames of -year and -year with rank are two best ones towards achieving the lowest RMSE ( root mean squared error ) value in matrix completion of H3N2 dataset ( Table 1 ) . The average RMSE from -year experiment is slightly better than that from -year experiment . Both the average RMSE for -year and -year experiment are better than that from the entire HI matrix . Thus , during matrix completion , a window of and a rank of will used . Similarly , our optimization method demonstrated that the window size of and the rank of are the best parameters for our simulation data . After the matrix completion step , we need to project the influenza antigens onto a two-dimensional ( 2D ) map . In order to obtain accurate global distances , we incorporate a temporal model in MDS based on the fact that the influenza antigens continue to evolve under the accumulating immune pressures of human population [9] . In order to evade the herd immunity , an influenza virus will most likely evolve into a strain with different antigenicity from recently circulating strains in human population . This intuition is mathematically incorporated in our temporal MDS model , where we assume that on the 2D cartography , influenza viruses tend to evolve along ( approximate ) straight-line segments during short time spans; that is , they tend to evolve in directions as far away from recently appeared viruses as possible . The detailed mathematical formula is presented in the Materials and Methods section . In HI tables , a Type II value is resulted from experimental limitation of HI assay and reflects a weak ( or low ) immunological reaction between a testing antigen/antiserum pair . Although this value is not as informative as a Type I value , it is more useful than a Type III value ( missing value ) . In particular , if a particular virus has type I values with a certain set of antibodies that show strong reactions , while another virus reacts weakly with the same set of antibodies ( resulting in type II values ) , then the global distance between their 2D cartography embeddings should be relatively large . A set of constraints on global distances can be derived from this observation . The details can be found in the Materials and Methods section . There are four parameters to be optimized in our temporal MDS model . We use -fold cross validations to select the optimal parameters that achieve the lowest RMSE while satisfying global distance constraints derived from Type II data . Our cross-validation results led to for the real data and for the simulation data . To demonstrate the potential impacts of Type II data ( low reactors ) and Type III data ( missing values ) on the influenza cartography , we performed experiments using simulated HI matrices containing antigens versus antibodies in which we know the ground-truth . Three simulated HI matrices were generated , where one was based on the distributions of H3N2 1968–2003 HI data: ( 1 ) HI matrix ( data absence ) with neither Type II nor Type III data; ( 2 ) HI matrix ( data absence , data structure: randomly distributed ) with Type III data but without Type II data; ( 3 ) HI matrix ( data absence , data structure: with a temporal data missing bias similar to H3N2 data as shown in Figure 2 ) with both Type II data and Type III data . The first HI matrix serves as the benchmark data ( ground truth ) . The second HI matrix is used to test the efficiency of standard matrix completion algorithms under the missing uniformly at random assumption . The third matrix is used to examine the efficacy of the temporal model in MDS . A more effective computational method would be expected to produce a cartography more similar to that of the benchmark matrix . Using these simulated HI matrices , we are able to compare the MC-MDS method proposed in this work to the original metric MDS method of [8] in terms of HI matrix completion and cartography construction accuracies . To assess whether MC-MDS and metric MDS can accurately recover the HI values in the HI data , we calculated the local RMSEs for the Type I data using -fold cross validation ( Table 2 ) . The experimental data were partitioned into parts , and each time we use parts for training and part for testing . The RMSE values were calculated using the Type I values in the testing part . Here we only use Type I values for RMSE calculation in order to be consistent with our real-data experiment , where we do not know the ground-truth corresponding to Type II and Type III data . The local RMSE values were for MC-MDS and for metric MDS , where the notation of is used . Since a typical matrix value is about , these local RMSE values indicate that both methods were able to recover HI values effectively . The small difference between the two means of MC-MDS and metric MDS is significantly smaller than the standard deviations . Hence they are statistically insignificant . However , we note that metric MDS has a larger standard deviation , which is consistent with our observation that it is less stable . The effectiveness of a cartography construction algorithm can be assessed using figures of merit that measure its robustness and correctness . The robustness of a method is determined by the correlation coefficient ( CC value ) that is calculated from the pairwise distances among antigens for every two independent runs . The correctness of cartography is measured by two values: the difference between the maximum distances ( MD value ) between any antigens in the benchmark cartography and that from the method being evaluated ( either MC-MDS or metric MDS ) ; the pairwise distance RMSEs ( PD value ) , calculated by measuring the difference between the pairwise distances among all antigens in the benchmark cartography and those from the method being evaluated . We performed independent runs , and the mean and standard deviation for each figure of merit can be found in Table 2 . As specified in the Materials and Methods section , the matrix completion method employed in this paper was Alternating Gradient Descent ( AGD ) . In Figure 3 , the ground-truth cartography is given in Figure 3a . Figure 3b shows a typical result when matrix entries are missing uniformly at random ( the second matrix generated in our simulation study ) , where the standard AGD method accurately reconstructed cartography since the resulting cartography is similar to that from the benchmark matrix . Figure 3c shows a typical result with temporally biased HI table ( the third matrix generated in our simulation study ) , where the cartography was constructed from a combination of AGD for matrix completion and the conventional MDS ( without temporal modeling ) for cartography generation . It shows that this combination is unable to accurately recover the cartography of the benchmark data since the global distances are incorrect . In comparison , the combination of AGD with temporal MDS , shown in Figure 3d , does achieve significantly more accurate global cartography . This experiment demonstrates the need to explicitly incorporate temporal modeling into the MDS step . Moreover , our experiment shows that cartographies generated by AGD and temporal MDS are stable . The CC value and PD value for the independent runs are and , respectively . The MD value for the independent runs is , which is close to the ground-truth value of in the benchmark cartography ( Figure 3a ) . For comparison , we implemented the metric MDS method of [8] and applied to the third HI matrix which was generated with temporally biased data type distributions . Our results indicate that the cartographies from metric MDS are less stable , with two typical runs given in Figure 3e and 3f . In the independent runs of metric MDS , the CC value and PD value for the independent runs are and . The MD value is . These numbers are significantly worse than the corresponding numbers from the MC-MDS method proposed in this work . We shall especially note that the metric MDS method tends to over-estimate the global distances in this stimulation study . Moreover , the large standard deviations in the results also indicate that metric MDS is not very stable . While these two methods achieve similar matrix completion accuracies , the reconstructed cartographies reveal a more significant difference . As we pointed out earlier , this is because the temporal bias ( of data type distribution ) in HI tables has stronger impact in the MDS step , especially when we compare global distances . Without temporal modeling , the accuracy of global distances between two points ( representing two viruses ) in the 2D cartography decays more rapidly when the two points become further apart in time . While this reduction of accuracy is an unavoidable limitation of the banded structure in HI tables ( Figure 1 ) that makes it harder to reliably compare points far away in time , a good temporal model can alleviate its impact , and thus increase the accuracy of the resulting cartography . Finally we summarize the main observations from this simulation study as follows . Both MC-MDS and metric MDS methods achieved similar accuracy in recovering HI values . This means that they achieve comparable performance in the matrix completion sub-task , which is less sensitive to the temporal bias problem in HI tables . However , without temporal modeling , the global distances among far away points in the reconstructed cartography become inaccurate . Therefore it is helpful to incorporate temporal modeling into the MDS step in order to reduce the temporal bias effect . The proposed MC-MDS framework ( with herd-immunity temporal model ) is effective in reducing the bias problem , and it leads to more accurate cartography . The metric MDS appears to be less stable and it generates less accurate cartographies because the method does not address the temporal bias problem . In the second experiment , we use MC-MDS to construct influenza cartography for H3N2 influenza A viruses from 1968 to 2003 using the HI datasets from Smith et al . [8] . The antigenic map is shown in Figure 4 . The scale of antigenic cartography is based on the antigenic distances from HI tables , e . g . each unit ( grid ) in the antigenic cartography represents of a 2-fold change in HI titres . These viruses are specifically labeled as eleven clusters ( HK68 , EN72 , VI75 , TX77 , BK79 , SI87 , BE89 , and BE92 , WU95 , SY97 , and FU02 ) . Our results indicate that the antigenic distance between HK68 and FU02 is approximately units . The resulting cartography can be compared to the published antigenic map in Smith et al . [8] . The overall trend in our results is similar to the cartography from Smith et al . [8] . However , there are two major differences: ( 1 ) The global distances in our cartography are smaller than those of Smith et al . [8] . For example Smith et al . [8] shows a distance of units between HK68 and FU02 . Although we have no ground truth for this data , we note that this discrepancy is consistent with our simulation study , where the metric MDS method also produces larger global distances . In that case , the metric MDS method over-estimated the global antigenic distance between A and J by units more than the true distance . ( 2 ) The local cartographies between some clusters are different . For instance , the distance between WU95 and BE89 from our method is larger than those shown in Smith et al . [8] . In order to examine which antigenic cartography is likely to be more accurate , we performed a small cartography for H3N2 HI data from 1987 to 1995 . Since the number of Type II data on the HI data from 1987 to 1995 is quite small , the effects of Type II on the antigenic cartography is minimal . Therefore , the cartography for the viruses between 1987 to 1995 using data from the limited span will not suffer much from the temporal bias problem discussed in the paper , and thus should be close to the true cartography . Our result shows that the distance between WU95 and BE89 should indeed be larger than that between BE95 and BE92 ( Figure 5 ) , and this is consistent with the local cartographies from MC-MDS . Similar to the simulated HI data experiments , we can assess the robustness of MC-MDS and metric MDS on the H3N2 data ( Table 2 ) . The best local RMSE was for MC-MDS and for metric MDS . Therefore there is no statistically significant difference in matrix completion quality . The CC values from the independent runs are and for MC-MDS and metric MDS , respectively . The MD value was for MC-MDS and for metric MDS . These numbers are consistent with the simulation study , showing again that MC-MDS is more stable for antigenic cartography construction . From the runs of metric MDS , we were not able to generate the exact cartography in Smith et al . [8] . One reason might be that the initial values we randomly chose were not exactly the same as those from [8] , which were not specified clearly from [8] . The source code of our implementation of the metric MDS method in [8] is available upon request . We shall point out that our implementation is strictly based on what was described in [8] . While we have spent great effort to ensure the correctness of our implementation , it is possible that there are undocumented improvements in the optimization algorithm used to solve the metric-MDS problem . In such case , their actual implementation might not suffer from the issues observed in our study . Nevertheless it is still useful for us to examine problems of the algorithm presented in their original paper , the underlying causes of these problems and their potential mathematical remedies . This is what this study tries to achieve .
We introduced a new computational framework for influenza antigenic cartography construction from HI datasets . This approach , which we refer to as MC-MDS , integrates two mathematical procedures: matrix completion and MDS projection ( with temporal modeling ) . Using the AGD matrix completion algorithm on HI datasets from 1968 to 2003 , we successfully identified the eleven reported clusters of antigenic variants that represent major antigenic drift events during these years . Thus , this method is useful in both influenza antigenic variant identification and influenza vaccine strain selection . Our results also demonstrated that MC-MDS is more robust and effective than our implementation of the metric MDS method [8] in influenza antigenic cartography construction . | Influenza antigenic cartography is an analogy of geographic cartography , and it projects influenza antigens into a two- or three-dimensional map through which we can visualize and measure the antigenic distances between influenza antigens as we visualize and measure geographic distances between the cities in a geographic cartography . Thus , influenza antigenic cartography can be utilized to identify influenza antigenic variants , and it is useful for influenza vaccine strain selection . Here we develop a new computational framework for constructing influenza antigenic cartography based on hemagglutination inhibition assay , a routine antigenic characterization method in influenza surveillance and vaccine strain selection . This method can be used for antigenic characterization in vaccine strain selection for both seasonal influenza and pandemic influenza . | [
"Abstract",
"Introduction",
"Results",
"Discussions"
] | [
"computer",
"science",
"immunology/immune",
"response",
"virology",
"evolutionary",
"biology/bioinformatics",
"computational",
"biology",
"mathematics/statistics"
] | 2010 | A Computational Framework for Influenza Antigenic Cartography |
A previous genome-wide association ( GWA ) meta-analysis of 12 , 386 PD cases and 21 , 026 controls conducted by the International Parkinson's Disease Genomics Consortium ( IPDGC ) discovered or confirmed 11 Parkinson's disease ( PD ) loci . This first analysis of the two-stage IPDGC study focused on the set of loci that passed genome-wide significance in the first stage GWA scan . However , the second stage genotyping array , the ImmunoChip , included a larger set of 1 , 920 SNPs selected on the basis of the GWA analysis . Here , we analyzed this set of 1 , 920 SNPs , and we identified five additional PD risk loci ( combined p<5×10−10 , PARK16/1q32 , STX1B/16p11 , FGF20/8p22 , STBD1/4q21 , and GPNMB/7p15 ) . Two of these five loci have been suggested by previous association studies ( PARK16/1q32 , FGF20/8p22 ) , and this study provides further support for these findings . Using a dataset of post-mortem brain samples assayed for gene expression ( n = 399 ) and methylation ( n = 292 ) , we identified methylation and expression changes associated with PD risk variants in PARK16/1q32 , GPNMB/7p15 , and STX1B/16p11 loci , hence suggesting potential molecular mechanisms and candidate genes at these risk loci .
Until the recent developments of high throughput genotyping and genome-wide association ( GWA ) studies , little was known of the genetics of typical Parkinson's disease ( PD ) . Studies of the genetic basis of familial forms of PD first identified rare highly penetrant mutations in LRKK2 [1] , [2] , PINK1 [3] , SNCA [4] , PARK2 [5] and PARK7 [6] . Following these findings , GWA scans for idiopathic PD identified SNCA and MAPT as unequivocal risk loci [7] , [8] , [9] , [10] , [11] as well as implicated BST1 [8] , GAK [12] , and HLA-DR [13] . Using sequence based imputation methods [14] , the meta-analysis of several GWA scans [7] , [9] , [10] , [11] conducted by the International Parkinson's Disease Genomics Consortium ( IPDGC ) identified and replicated five new loci: ACMSD , STK39 , MCCC1/LAMP3 , SYT11 , and CCDC62/HIP1R [15] and confirmed association at SNCA , LRRK2 , MAPT , BST1 , GAK and HLA-DR [15] . We conducted a two-stage association study . Combining stage 1 and stage 2 , the data consist of 12 , 386 PD cases and 21 , 026 controls genotyped using a variety of platforms ( Table 1 ) . Stage 1 used genome-wide genotyping arrays and our initial analysis [15] focused on the subset of SNPs that passed genome-wide significance in stage 1 . For stage 2 genotyping , we used a custom content Illumina iSelect array , the ImmunoChip and additional GWAS typing as previously described [15] . The primary content of the ImmunoChip data focuses on autoimmune disorders but , as part of a collaborative agreement with the Wellcome Trust Case Control Consortium 2 , we included 1 , 920 ImmunoChip SNPs on the basis of the stage 1 GWA PD results . Here , we report the combined analysis for this full set of 1 , 920 SNPs . This step1+2 analysis identified seven new loci that passed genome-wide significance in the meta-analysis . During the process of analyzing these data and preparing for publication , we became aware that another group was also preparing a large independent GWA scan in PD for publication ( Do et al , submitted ) . Following discussion with this group we agreed to cross validate the top hits from each study by exchanging summary statistics for this small number of loci . To provide further insights into the molecular function of these associated variants , we tested risk alleles at these loci for correlation with the expression of physically close gene ( expression quantitative trait locus , eQTL ) and the methylation status ( methQTL ) of proximal DNA CpG sites in a dataset of 399 control frontal cortex and cerebellar tissue samples extracted post-mortem from individuals without a history of neurological disorders .
In addition to eleven loci that passed genome-wide significance in stage 1 [15] , we identified over 100 regions of interest defined as 10 kb windows containing at least one SNP associated at p<10−3 . We submitted the most associated SNP in each region for probe design and follow-up genotyping using the ImmunoChip platform . For each region of interest , we also added four SNPs in high level of linkage disequilibrium ( LD ) to provide redundancy where the most associated SNP would not pass the Illumina probe design step or the assay for that SNP would fail . To complete the array design we also added all non-synonymous dbSNPs located in known PD associated regions [1] , [2] , [3] , [4] , [5] , [6] . Out of these 2 , 400 submitted SNPs , 1 , 920 passed QC and were included in the final array design . For these 1 , 920 SNPs we combined stage 1 and stage 2 associated data in a meta-analysis of 12 , 386 cases and 21 , 026 controls ( Table 1 ) from the IPDGC . We exchanged summary statistics for these most significant hits with an additional large , case-control replication dataset ( 3 , 426 PD cases and 29 , 624 controls ) in an attempt to demonstrate independent replication . On the basis of stage 1+2 results , seven new SNPs passed our defined genome-wide significance threshold ( p<5×10−8 , Table 2 and Figure 1 ) . These loci are either novel or the previous evidence of association was not entirely convincing in individuals of European descent . We combined these results with the independent replication . Five of these seven loci replicated and showed strong combined evidence of PD association ( p<10−10 overall ) . Taking either the nearest gene ( or the strongest candidate when available ) to designate these regions , these five loci are 1q32/PARK16 [7] , 4q21/STBD1 , 7p15/GPNMB , 8p22/FGF20 [16] and 16p11/STX1B . rs708723/1q32 has been previously reported as PD associated ( PARK16 , [7] , [8] ) but this SNP lacked the unequivocal evidence of association in European samples ( p = 9 . 47×10−10 in stage 2 only ) . To understand the potential biological consequences of risk variation at this locus we tested whether rs708723 was correlated with either gene expression or DNA methylation status of proximal transcripts or CpG sites respectively ( Table 3 ) . We found correlations with the expression of NUCKS1 ( p = 1 . 8×10−7 ) and RAB7L1 ( p = 7 . 2×10−4 ) . We also found correlations with the methylation state of CpG sites located in the FLJ3269 gene ( p = 3 . 9×10−22 ) . In the case of 16p11/STX1B , the proximal gene to the most associated SNP rs4889603 is SETD1A . However , STX1B is located 18 kb upstream of rs4889603 and is a more plausible PD candidate gene [17] owing to its synaptic receptor function . We therefore used this gene to designate this region . Our methQTL/eQTL dataset identified a correlation between the rs4889603 risk allele and increased methylation of a CpG dinucleotide in STX1B ( Table 3 ) . The SNP rs591323 in the 8p22 region is located ∼150 kb downstream of the FGF20 gene ( NCBI build 36 . 3 ) , for which association with PD has been suggested previously in familial PD samples [16] , [18] but which remained controversial [19] . Our findings provide further support for a PD association at this locus , but again , whether the functionally affected transcript is FGF20 or not remains unclear . The regions 4q21/STBD1 and 7p15/GPNMD have not been previously implicated in PD etiology . We found that the risk allele of rs156429 , the most associated SNP in the 7p15 region , is associated in our eQTL dataset with decreased expression of the proximal transcript encoded by NUPL2 ( Table 3 ) . The same risk allele is also associated with increased methylation of multiple CpG sites proximal to GPNMB itself ( Table 3 ) . Neither of these regions contains an obvious candidate gene . Two additional loci ( 3q26/NMD3 and 8q21/MMP16 ) showed strong evidence of association in stage 1 and 2 but were not disease associated in the Do et al dataset . Further replication is required to clarify the role of variation at these loci in risk for PD . The strongly associated G2019S variant in the LRRK2 gene [20] was included in the Immunochip design and we replicated the published association: control frequency: 0 . 045% case frequency 0 . 61% , estimated odds ratio: 13 . 5 with 95% confidence interval: 5 . 5–43 . However , the case collections have been partially screened for this variant therefore its frequency in cases and the odds ratio is likely to be underestimated . The ImmunoChip array design provides some power to detect whether multiple distinct association signals exist at individual loci . Indeed , if a SNP showed an independent and sufficiently strong association in stage 1 , it would have been included in stage 2 provided that it was not located in the same 10 kb window as the primary SNP in the region . There is precedent for this in PD , with the previous identification of independent risk signals at the SNCA locus [11] . We therefore used the Immunochip data to test whether any of the seven loci in Table 2 showed some evidence of more than one independent signal . None of these seven loci showed any association ( p>0 . 01 ) after conditioning on the main SNP in the region . In contrast , after conditioning on the most associated SNPs rs356182 in the SNCA region , several SNPs remained convincingly associated ( p = 9 . 7×10−8 for rs2245801 being the most significant ) . Lastly , we performed a risk profile analysis to investigate the power to discriminate cases and controls on the basis of the 16 confirmed common associated variants ( Table 4 ) . For each locus , we estimated the odds ratio on the basis of stage 1 data and we applied these estimates to compute for each individual in the ImmunoChip cohort a combined risk score . Solely based on these 16 common variants , and therefore not considering rare highly penetrant variants such as G2019S in LRKK2 [20] , we found that individuals in the top quintile of the risk score have an estimated three-fold increase in PD risk compared to individuals in the bottom quintile ( Table 4 ) . We note however that the effect size of several of these associated variants could be over-estimated ( an effect known as winner's curse , see [21] ) but given the consistent estimates of odds ratio across studies ( Table 4 ) we expect this bias to be minimal .
The combination of GWA scans and imputation methods in large cohorts of PD cases and controls has enabled us to identify five PD associated loci in addition to the 11 previously reported by us . Two of these loci ( 1q32/PARK16 , 8p22/FGF20 ) implicate regions that had been previously associated with PD risk [8] , [16] . The 1q32/PARK16 showed convincing evidence of association in the Japanese population [8] but until now the association P-value had not passed a stringent genome-wide significance threshold in samples of European descent [7] . The 8p22/FGF20 locus had been previously reported in a study of familial PD [16] and we provide the first evidence of association in a case-control study . The remaining three loci ( STX1B/16p11 , STBD1/4q21 and GPNMB/7p15 ) are new . Adding the eleven previously reported common variants [15] to the five convincingly associated loci identified in this study , common variants at 16 loci have now been associated with PD . Controlling for the risk score based on the 11 SNPs previously identified [15] in the risk profile analysis ( Table 4 ) , the addition of these five new loci provides a modest but significant ( p = 2 . 2×10−3 ) improvement of our ability to discriminate PD cases from controls . Combining eQTL/methylation and case-control data implicates potential mechanisms which could explain the increased PD risk associated some of these variants . In particular , the strong eQTL in the 1q32/PARK16 region with the RAB7L1 and NUCKS1 genes ( Table 3 ) suggests that either one of these genes could be the biological effector of this risk locus . However , existing data show that eQTLs are widespread and this co-localization could be the result of chance alone [22] . Additional fine-mapping work will be required to assess whether the expression and case-control data are indeed fully consistent . While we are unable to unequivocally pinpoint the causative genes underlying these associations , their known biological function can suggest likely candidates . At the 1q32/PARK16 loci our association and eQTL data indicate that RAB7L1 and NUCKS1 are the best candidates . The former is a GTP-binding protein that plays an important role in the regulation of exocytotic and endocytotic pathways [23] . Exocytosis is relevant for PD for two main reasons: firstly , since dopaminergic neurotransmission is mediated by the vesicular release of dopamine , i . e . dopamine exocytosis [24] , and secondly because it has been shown that alpha-synuclein knock-out mice develop vesicle abnormalities [25] , thus providing a potential direct link between genetic variability in the gene and a biological pathway involved in the disease . Less is known regarding NUCKS1; it has been described to be a nuclear protein , containing casein kinase II and cyclin-dependant kinases phosphorylation sites and to be highly expressed in the cardiac muscle [26]; but an involvement in PD pathogenesis has yet to be suggested . At the 16p11/STX1B locus , notwithstanding the fact that other genes are in the associated region , STX1B is the most plausible candidate . It has been previously shown to be directly implicated in the process of calcium-dependent synaptic transmission in rat brain [17] , having been suggested to play a role in the excitatory pathway of synaptic transmission . Since parkin , encoded by PARK2 , negatively regulates the number and strength of excitatory synapses [27] , it makes STX1B a very interesting candidate from a biologic perspective . FGF20 at 8p22 has been suggested to be involved in PD [16] , albeit negative results in smaller cohorts have followed the original finding [28] . FGF20 is a neurotrophic factor that exerts strong neurotrophic properties within brain tissue , and regulates central nervous development and function [29] . It is preferentially expressed in the substantia nigra [30] , and it has been reported to be involved in dopaminergic neurons survival [30] . The ImmunoChip data provide limited resolution for the detection of multiple independent association signals in these regions . A previous study [31] reported some evidence of allelic heterogeneity at the 1q32/PARK16 locus but the ImmunoChip data do not support this result . A previous study [11] also reported two independent associations at the 4q22/SNCA locus and our data are consistent with this scenario . However , the newly reported secondary association ( rs2245801 ) is in low LD ( r2 = 0 . 21 ) with rs2301134 , the SNP reported in [11] as an independent association . Taken together , these findings suggest that at least three independent associations exist at SNCA/4q22 . A more exhaustive fine-mapping analysis using either sequencing of large cohorts or targeted genotyping arrays will also be required to fully explore this locus . As yet , we do not know which of the variants and which genes within each region are exerting the pathogenic effect . We cannot exclude that some of the currently reported variants are in fact tagging high penetrance , but rare , mutations [32] . Nevertheless , the successful identification of these 16 risk loci further demonstrates the power of the GWA study design , even in the context of disorders like PD that have a complex genetic component . We therefore expect that further and larger association analyses , perhaps using dedicated high-throughput genotyping arrays like the ImmunoChip , will continue to yield new insights into PD etiology .
Participating studies were either genotyped using the ImmunoChip as part of a collaborative agreement with the ImmunoChip Consortium , or as part of previous GWA studies provided by members of the IPDGC or freely available from dbGaP [7] , [9] , [10] , [11] . Genotyping of the UK cases using the Immunochip was undertaken by the WTCCC2 at the Wellcome Trust Sanger Institute which also genotyped the UK control samples . The constituent studies comprising the IPDGC have been described in detail elsewhere [15] , although a summary of individual study quality control is available as part of Table S1 . In brief all studies followed relatively uniform quality control procedures such as: minimum call rate per sample of 95% , mandatory concordance between self-reported and X-chromosome-heterogeneity estimated sex , exclusion of SNPs with greater than 5% missingness , Hardy Weinberg equilibrium p-values at a minimum of 10−7 , minor allele frequencies at a minimum of 1% , exclusion of first degree relatives , and the exclusion of ancestry outliers based on either principal components or multidimensional scaling analyses using either PLINK [33] or EIGENSTRAT [34] to remove non-European ancestry samples . All GWAS studies utilized in this analysis ( and in the QTL analyses ) were imputed using MACHv1 . 0 . 16 [14] to conduct a two-stage imputation based on the August 2009 haplotypes from initial low coverage sequencing of 112 European ancestry samples in the 1000 Genomes Project [35] , filtering the data for a minimum imputation quality of ( RSQR>0 . 3 ) [14] . Logistic regression models were utilized to quantify associations with PD incorporating allele dosages as the primary predictor of disease . Imputed data was analyzed using MACH2DAT , and genotyped SNPs were analyzed using PLINK . All models were adjusted for covariates of components 1 and 2 from either principal components or multidimensional scaling analyses to account for population substructure and stochastic genotypic variation ( except in the UK-GWAS data which were not adjusted for population substructure ) . Single SNP test statistics were combined across datasets using a score test methodology , essentially assuming equal odds ratio across cohorts . In addition , fixed and random effects meta-analyses were implemented in R ( version 2 . 11 ) to confirm that the score test approximation does not affect the interpretation of the results . We also tested the relevant SNPs heterogeneity across cohorts and no significant heterogeneity was detected ( Table S2 ) . We communicated to our colleagues in charge of the independent study ( Do et al ) the seven SNPs listed in Table 2 . For this subset of SNPs they selected the marker with the highest r2 value on their genotyping platform and provided us with the following summary statistics: odds ratio , direction of effect , standard error for the estimated odds ratio and one degree-of-freedom trend test P-value . Quantitative trait analyses were conducted to infer effects of risk SNPs on proximal CpG methylation and gene expression . For the five replicated SNP associations ( Table 2 ) , all available CpG probes and expression probes within +/−1 MB of the target SNP were investigated as candidate QTL associations in frontal cortex and cerebellar tissue samples . 399 samples were assayed for genome-wide gene expression on Illumina HumanHT-12 v3 Expression Beadchips and 292 samples were assayed using Infinium HumanMethylation27 Beadchips , both per manufacturer's protocols in each brain region . A more in depth description of the sample series comprising the QTL analyses , relevant laboratory procedures and quality requirements may be found in [15] . The QTL analysis utilized multivariate linear regression models to estimate effects of allele dosages per SNP on expression and methylation levels adjusted for covariates of age at death , gender , the first 2 component vectors from multi-dimensional scaling , post mortem interval ( PMI ) , brain bank from where the samples were provided and in which preparation/hybridization batch the samples were processed . A total of 670 candidate QTL associations were tested: 87 expression QTLs in the cerebellum samples , 85 expression QTLs in the frontal cortex samples , 249 methylation QTLs in the cerebellum samples and 249 methylation QTLs in the frontal cortex samples . Multiple test correction was undertaken using false discovery rate adjusted p-values<0 . 05 to dictate significance , with the p-value adjustment undertaken in each series separately , stratified by brain region and assay . A complete list of all QTL associations tested is included in Table S3 . | This paper describes the largest case-control analysis of Parkinson's disease to date , with a combined sample set of over 12 , 000 cases and 21 , 000 controls . After combining our findings with an independent replication dataset of more than 3 , 000 cases and 29 , 000 controls , we found five additional PD risk loci in addition to the 11 loci previously identified in earlier consortium efforts . This successful study further demonstrates the power of the GWA scan experimental design to find new loci contributing to disease risk , even in the context of complex disorders like Parkinson's disease . These new findings provide insights into the etiology of PD and will promote a better understanding of its pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"genetics",
"biology",
"genetics",
"of",
"disease",
"neuroscience",
"genetics",
"and",
"genomics"
] | 2011 | A Two-Stage Meta-Analysis Identifies Several New Loci for Parkinson's Disease |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.